Search

CN-122026792-A - Photovoltaic steam self-cleaning system based on multi-source sensing and waste heat cooperation and control method

CN122026792ACN 122026792 ACN122026792 ACN 122026792ACN-122026792-A

Abstract

The invention belongs to the technical field of operation and maintenance of photovoltaic power generation systems, and discloses a photovoltaic steam self-cleaning system based on multi-source sensing and waste heat cooperation, which comprises: the intelligent automatic spray gun comprises an inverter efficiency monitoring module, a visual AI stain identification module, an intelligent decision controller, a multi-heat-source cooperative system, a steam generation and low-pressure safety system, a thermal shock protection water mixing device, an automatic spray gun and a cleaning execution unit. The system integrates the inverter efficiency monitoring and vision AI fusion recognition, combines the attenuation rate to accurately distinguish stains and shadows, improves the accuracy of cleaning decisions, integrates multi-parameter safety interlocking of temperature, humidity, wind speed and the like, eliminates the risks of severe weather operation, innovatively utilizes heating ventilation, energy storage and inverter waste heat to construct a multi-heat source synergistic system, thoroughly gets rid of the dependence of traditional electric heating through MPC algorithm scheduling, and effectively limits the temperature rise of a panel through mixed water cooling and closed loop temperature control, thereby fundamentally avoiding thermal shock and prolonging the service life of components.

Inventors

  • YANG HONGWEI

Assignees

  • 比塔（上海）数据科技有限公司

Dates

Publication Date
20260512
Application Date
20260211

Claims (10)

  1. 1. A photovoltaic steam self-cleaning system based on multi-source perception AND waste heat cooperation is characterized by comprising an inverter efficiency monitoring module, a vision AI (advanced technology attachment) spot recognition module, an intelligent decision controller, a multi-heat source cooperation system, a steam generation AND low-pressure safety system, a thermal shock protection water mixing device, an automatic spray gun AND a cleaning execution unit, wherein the inverter efficiency monitoring module reads inverter data in real time through a Modbus TCP interface, calculates theoretical power generation, efficiency attenuation AND attenuation time derivative to distinguish shadows from spots, the vision AI module collects component images AND detects the spots through a YOLOv/v 12 model, the coverage rate AND the thermodynamic diagram are output, the intelligent decision controller triggers cleaning through multi-condition AND logic, the multi-heat source cooperation system utilizes heating AND ventilation condensation heat, energy storage heat dissipation AND the inverter heat dissipation to generate steam, the thermal shock protection device avoids thermal damage through water mixing AND closed-loop control, AND the automatic spray gun is accurately cleaned according to the thermodynamic diagram.
  2. 2. The photovoltaic steam self-cleaning system based on multi-source sensing and waste heat cooperation as claimed in claim 1, wherein the theoretical generation power calculation formula of the inverter efficiency monitoring module is as follows In the following The peak power of the photovoltaic is rated, For the irradiance of the horizontal plane, For a temperature coefficient of 0.0045, In order to measure the temperature of the component, The value range of the correction coefficient for the inclined plane incidence angle is [0.85,1.0].
  3. 3. The photovoltaic steam self-cleaning system based on multi-source perception and waste heat synergy according to claim 1 is characterized in that an IP67 protection grade, cameras with resolution of more than or equal to 2MP and FOV of more than or equal to 90 degrees are mounted on a visual AI spot recognition module, one frame of image is collected every 30 seconds, a model is input after graying and normalization pretreatment, mAP@0.5 is more than or equal to 95%, inference delay is less than 200 ms/frame, spot coverage rate C is calculated to be +/-5%, and the spot coverage rate C is divided into three grades of normal, moderate and severe according to C <10%, 10% -30% and more than or equal to 30%.
  4. 4. The photovoltaic steam self-cleaning system based on multi-source perception AND waste heat synergy according to claim 1, wherein the multi-condition AND logic of the intelligent decision controller comprises AI stain coverage rate C >10%, efficiency attenuation percentage DP% >5%, attenuation time derivative dDP/dt <1.0, environment temperature T less than or equal to 35 ℃, wind speed V less than or equal to 8m/s, relative humidity RH less than or equal to 85%, AND time from last cleaning is more than 4 hours, AND when all conditions are met simultaneously, a trigger cleaning signal is output, otherwise a negative signal is returned or a waiting queue is added.
  5. 5. The photovoltaic steam self-cleaning system based on multi-source sensing and waste heat synergy according to claim 1, wherein the multi-source sensing and waste heat synergy system comprises three heat sources, namely heating condensation heat Q_HVAC, power is 15-25kW, energy storage heat dissipation Q_ESS, power is 3-6kW, inverter heat dissipation Q_INV and power is 1-4kW, the multi-source sensing and waste heat synergy system adopts an MPC algorithm to conduct heat source scheduling, 10 minutes is taken as a prediction window, optimal mixing proportion is solved through quadratic programming, and optimization targets are that heat waste is minimized, and heat exchange efficiency is more than or equal to 85%.
  6. 6. The photovoltaic steam self-cleaning system based on the multi-source sensing and waste heat synergy is characterized in that parameters of steam generated by the steam generation and low-pressure safety system are 0.1-0.2MPa, 100-120 ℃ and 5-10kg/h in flow, the steam generation and low-pressure safety system is provided with a pressure gauge, a safety valve with the pressure of 0.25MPa and a quality monitoring system, and the pressure is automatically released when the pressure exceeds the pressure, so that the pressure meets the low-pressure safety standard.
  7. 7. The photovoltaic steam self-cleaning system based on multi-source sensing and waste heat cooperation, which is disclosed in claim 1, is characterized in that the thermal shock protection water mixing device adopts an electric three-way water mixing valve, steam with the temperature of more than 100 ℃ is mixed with cooling liquid with the temperature of about 25 ℃, the temperature of the steam after mixing is less than or equal to 60 ℃ and the temperature difference between the steam and a cold photovoltaic panel is less than or equal to 70 ℃, closed loop control adopts first derivative adjustment, the temperature rise rate of the panel is less than or equal to 5 ℃ per minute, the absolute temperature rise is less than or equal to 20 ℃, and the cleaning and the cooling mode are immediately stopped and started if any limit is exceeded.
  8. 8. The photovoltaic steam self-cleaning system based on multi-source sensing and waste heat coordination according to claim 1, wherein the automatic spray gun and the cleaning execution unit are provided with an automatic scanning mechanism, a high coverage area is cleaned preferentially according to a stain distribution thermodynamic diagram of an 8X 8 grid, the opening quantity of the spray gun is adjusted dynamically by 30% -70% according to the temperature rise rate, the cleaning period is 8-12 minutes, and the automatic spray gun and the cleaning execution unit are switched to a cooling mode automatically after completion.
  9. 9. A photovoltaic steam control method based on multi-source sensing and waste heat cooperation is characterized by comprising the following specific steps: collecting data of an inverter, environment and a heat source every minute, and collecting an image of a component every 30 seconds; Calculating efficiency attenuation and derivative, distinguishing shadows from stains, generating a stain coverage rate and a thermodynamic diagram, and reporting all data to a AIHUB database; step three, decision judgment, namely judging whether the five conditions in claim 4 are met, triggering cleaning if the conditions are met, and returning to a parameter acquisition stage to wait for the next period if the conditions are not met; step four, heat source scheduling, namely determining heat source mixing ratio through an MPC algorithm to generate up-to-standard steam; step five, cleaning and executing, namely recording the initial temperature of the photovoltaic panel The spray gun is started to spray mixed steam, and closed loop control is carried out once per second, wherein the current temperature of the panel is detected, if the current temperature exceeds the current temperature The cleaning is stopped in emergency at +20 ℃, the temperature rise rate is calculated, the number of spray guns is reduced by 30% if the temperature rise rate exceeds 5 ℃ per minute, the steam temperature is detected, and the mixing ratio of cooling liquid is increased if the temperature rise rate exceeds 60 ℃; step six, cooling, namely closing steam supply after cleaning is finished, and starting a cooling liquid circulating pump, when the temperature of the panel is reduced to be within +5 ℃ of the ambient temperature and the temperature of the cooling liquid is less than 30 ℃, closing the cooling pump, recording a cleaning time stamp and a power generation recovery effect, and reporting AIHUB; And step seven, waiting for circulation, namely setting a next cleaning interval, and returning to a parameter acquisition stage to enter the next circulation.
  10. 10. The photovoltaic steam control method based on the cooperation of multi-source sensing and waste heat according to claim 9, wherein in the fourth step, the steam generation is realized by heating the mixed liquid preheated by a plurality of heat sources for 5-10 minutes in a sectional temperature control mode.

Description

Photovoltaic steam self-cleaning system based on multi-source sensing and waste heat cooperation and control method Technical Field The invention belongs to the technical field of operation and maintenance of photovoltaic power generation systems, and particularly relates to a photovoltaic steam self-cleaning system based on multi-source sensing and waste heat cooperation and a control method. Background With the wide deployment of distributed photovoltaic power stations, the problem of power generation efficiency attenuation caused by accumulation of dirt such as dust, bird droppings, snow and the like on the surface of a photovoltaic module is increasingly prominent, and periodic cleaning is a key measure for guaranteeing power generation benefits of a system, and at present, the following cleaning schemes mainly exist in the industry, but have respective limitations: 1. traditional timing or manual cleaning modes: This mode typically performs a wash based on a fixed schedule (e.g., weekly or monthly), or relies on manual post-field inspection decisions; The intelligent water purifier has the main defects of low intelligent degree ①, incapability of accurately sensing the actual severity degree of stains and environmental influence, easiness in causing excessive cleaning or insufficient cleaning, waste of water and electric resources, loss of generated energy caused by the latter, no guarantee on ② safety, possibility of forced operation under severe weather such as strong wind, low temperature, high humidity and the like, safety risks such as bracket shaking, pipeline frost cracking, personnel high-altitude operation and the like, high ③ operation and maintenance cost, high manual cleaning labor cost and low efficiency; 2. Automatic cleaning scheme based on single sensor: The prior art (such as CN114897918B, CN 117593675B) mainly uses visual images to identify stains, or uses a single power generation efficiency sensor to monitor power decay to trigger cleaning; The method has the main defects that ① identification accuracy is poor, single vision identification is easy to misjudge under backlight and shadow conditions, a single efficiency sensor cannot distinguish power fluctuation caused by instantaneous shadow shielding and power loss caused by long-term soil accumulation, so that false triggering or missed triggering is caused, ② lacks environment safety interlocking, cleaning triggering logic does not integrate key environment parameters such as temperature, wind speed, humidity and the like, cannot be automatically disabled under extreme weather, equipment damage risk is high, ③ cleaning mode is limited, a mechanical brush or a high-pressure water gun is adopted, cleaning effect is limited, components are possibly scratched, and the cleaning effect is not matched with a building energy system; 3. Existing solutions for steam cleaning systems: the electric heating type steam system directly uses the power generated by the photovoltaic system to heat and generate steam, has huge energy consumption, generally consumes 25% -40% of the self power generation of the system, seriously counteracts the power generation gain caused by cleaning, has poor economical efficiency and does not accord with the zero carbon concept; The engine waste heat driving scheme (such as CN 104096691B) is characterized in that the engine waste heat driving scheme utilizes industrial engine waste heat to generate steam, has the main defects that ① is limited in application scene, is only suitable for industrial power stations provided with large engines, cannot be suitable for distributed photovoltaic scenes such as industrial and commercial roofs, park buildings and the like which occupy the main stream of the market, ② lacks intelligent control, is mainly triggered by timing or manual operation, does not solve the problems of accurate recognition and safety triggering of stains, and has ③ thermal shock protection loss, namely does not consider the thermal shock risk caused by huge temperature difference (delta T can reach more than 120 ℃) between high-temperature steam and low-temperature photovoltaic glass panels, and can possibly cause glass breakage; 4. the existing scheme of building waste heat recovery: The prior art (such as CN 208475598U) mainly focuses on heat recovery of air conditioner condensed water, is used for internal circulation or flushing of a building, but does not establish effective cooperation with a photovoltaic cleaning system, has incomplete heat utilization chains and has seasonal limitation (such as no heat source when a heating and ventilation system is closed in winter); 5. Existing schemes for thermal management of energy storage systems: The prior art does not combine the stable low-grade heat source with the photovoltaic cleaning requirement, and cannot realize the cascade utilization of energy sources; in summary, the existing photovoltaic cleaning technology has common problems that decision er