CN-121980698-A - Cooling system failure judging method of fan frequency converter

CN121980698ACN 121980698 ACN121980698 ACN 121980698ACN-121980698-A

Abstract

The invention discloses a cooling system failure judging method of a fan frequency converter, which can more accurately identify cooling system faults, reduce false alarms and improve system reliability and operation efficiency by introducing a model attenuation coefficient and various prediction deviation models. The invention considers the system degradation caused by the increase of the service life of the fan, and the fault judging method can adapt to the characteristic change of different operation stages of the fan by dynamically adjusting the attenuation coefficient, thereby enhancing the robustness of the system, and even under the conditions of longer service life of the fan and reduced performance of the cooling system, the running state of the cooling system can be accurately judged, and the influence on the normal power generation of the fan caused by false alarm is avoided. The number and the duration of unplanned shutdown are effectively reduced, the operation efficiency of the wind power generation system is improved, and remarkable economic benefits are brought to energy companies.

Inventors

SONG YUEWEN
TANG RUI
ZHANG YUZHOU
HU XUDONG
Man Zhilin
TANG QING

Assignees

湖南省国智云科技有限公司

Dates

Publication Date: 20260505
Application Date: 20260105

Claims (9)

1. A cooling system failure judging method of a fan frequency converter is characterized by comprising the following steps: S1, acquiring environmental parameters in a specific area within 6-12 months and all fan operation parameters of the same model, wherein the environmental parameters at least comprise parameters of environmental temperature, wind speed and power generation; s2, based on S1, acquiring historical data about environmental parameters and fan running states in the area in the period of 6-12 months, uniformly modeling all fans of the same model, and constructing a model base prediction model delta P_M_B of the pressure difference of all fan frequency converter cooling systems of the same model; S3, based on the S2, testing and calculating a model current prediction model delta P_M_C and a model basic prediction model delta P_M_B to obtain an attenuation coefficient eta; S4, acquiring environmental parameters in the last 6-12 months in the same area as the step S1 and all fan operation parameters of the same model, modeling each fan of the same model, and constructing a current-stage fan prediction model delta P_M_Ci of the pressure difference of all fan frequency converter cooling systems of the same model; S5, based on S4, comparing the current cooling system pressure difference value of the fan frequency converter with a preset safety threshold T, and constructing a safety threshold judgment model, wherein if the current cooling system pressure difference value of the fan frequency converter is within a safety range, the threshold judgment model result takes a value of 0, and if the current cooling system pressure difference value of the fan frequency converter is not within the safety range, the result takes a value of 1, and the threshold judgment model result is represented by a; S6, based on the S5, calculating according to the current cooling system pressure difference value of the fan frequency converter and the predicted value of the current fan prediction model delta P_M_Ci, and constructing an individual deviation judgment model; s7, based on the S5, calculating according to the current cooling system pressure difference value of the fan frequency converter and the predicted value of the current fan prediction model delta P_M_Ci, and constructing a model deviation judgment model; S8, based on S6 and S7, constructing a comprehensive judgment model f according to the attenuation coefficient eta, the threshold judgment model result, the individual deviation judgment model result and the model deviation judgment model result to obtain a comprehensive judgment coefficient Z, and triggering a cooling system failure alarm if the calculated Z value is larger than a set threshold T, otherwise, considering that the fan can still normally operate.
2. The method for judging failure of a cooling system of a fan frequency converter according to claim 1, wherein in the step S1, a concrete construction flow of the model-in-period prediction model Δp_m_c is as follows: And the model current prediction model delta P_M_C of the model is constructed by adopting a least square method by taking the environmental temperature, wind speed and power generation power of 6-12 months recently as parameters and the corresponding pressure difference delta P of a cooling system of a fan frequency converter as a target value.
3. The method for judging failure of a cooling system of a fan frequency converter according to claim 2, wherein in the step S2, a concrete construction flow of the model base prediction model Δp_m_b is as follows: and setting the number of fans as n for unified modeling aiming at all fans with the same model in a specific area, selecting the ambient temperature, wind speed and power generation of the fans which are put into operation for 6-12 months as parameters, taking the pressure difference delta P of a corresponding fan frequency converter cooling system as a target value, and constructing a model base prediction model delta P_M_B of the model by adopting a least square method.
4. The method for determining failure of a cooling system of a fan frequency converter according to claim 3, wherein in the step S3, a specific calculation process of the attenuation coefficient η is as follows: For a current model of fans in a specific stator area, at least parameters of typical ambient temperature (theta 1, theta 2, theta M), wind speed (v 1, v2, vm), generation power (P1, P2, pm) are selected, and a pressure difference delta p_c (delta p_c1, delta p_c2, delta p_cm) and a pressure difference delta p_b (delta p_b1, delta p_b2) of the cooling system under the model typical parameters are calculated by using a delta p_m_c model and a delta p_m_b model respectively; η=α*(ΔP_C1/ΔP_B1)+α*(ΔP_C2/ΔP_B2) + ... ... + α*(ΔP_Cm/ΔP_Bm); Wherein alpha is a weight coefficient and takes a value of 1/m.
5. The method for judging failure of a cooling system of a fan frequency converter according to claim 4, wherein in the step S4, a specific construction flow of the current fan prediction model Δp_m_ci is as follows: Setting the number of fans as n for all fans with the same model in a specific area, respectively modeling, selecting the environmental temperature, wind speed and power generation power of 6-12 months recently as parameters, taking the pressure difference delta P of a corresponding fan frequency converter cooling system as a target value, and constructing fan prediction models delta P_M_C1, delta P_M_C2 and delta P_M_ C3. of n fans in the specific area by adopting a least square method.
6. The method for determining failure of a cooling system of a fan inverter according to claim 5, wherein in step S6, the specific content of the individual deviation determination model is as follows: According to a differential pressure prediction result delta Pi_pred obtained by specific calculation of a current fan prediction model delta P_M_Ci, if the actual differential pressure value delta Pi of the current fan is within a confidence interval of the differential pressure prediction model result, the individual deviation judgment model judgment result is 0, otherwise, the result takes a value of 1, and the individual deviation judgment model result is represented by b; the confidence interval is delta Pi_Pred+/-2 SE, wherein SE is the residual standard deviation, and the calculation formula of SE is as follows: 。
7. the method for judging failure of a cooling system of a fan inverter according to claim 6, wherein in step S7, the model deviation judging model has the following specific contents: calculating to obtain all fan differential pressure prediction results delta P_pred (delta P1_pred, delta P2_pred.) of the model by a current fan prediction model delta P_M_Ci; if the actual differential pressure value Δpi of the current blower is within the range of Δp_pred as a result of the differential pressure prediction model, that is, Δpi is within Δp1_pred, Δp2_pred. If the model deviation judging model is not an abnormal value, the model deviation judging model judging result is 0, otherwise, the result is 1, and the model deviation judging model result is expressed by c.
8. The method for determining failure of a cooling system of a fan inverter according to claim 7, wherein in step S8, the specific content of the comprehensive determination model f is as follows: Constructing a comprehensive judgment model f (eta, a, b and c) according to the attenuation coefficient eta, the threshold judgment model result, the individual deviation judgment model result and the model deviation judgment model result to obtain a comprehensive judgment coefficient Z, namely Z=f (eta, a, b and c), wherein the calculation formula of Z is as follows: 。
9. the method for judging failure of a cooling system of a fan frequency converter according to claim 6 or 7, wherein the specific calculation content of the least square method in the step S1, the step S2 and the step S4 is as follows: let (x, Δp) be a set of observables, x= [ θ, v, P ] T, the target pressure difference Δp satisfies the following theoretical function: ΔP = f(x,ω) wherein ω= [ ω1, ω2, ] T is the parameter to be fitted; In order to find an optimal estimate of the parameter ω of the function f (x, ω), for a given m sets of observations (xi, yi) (i=1, 2..once., m), the objective function is solved ; To optimize the fit, the objective function is calculated Minimum valued parameter i (i = 1,2,...,n)。

Description

Cooling system failure judging method of fan frequency converter Technical Field The invention relates to the technical field of fans, in particular to a failure judging method for a cooling system of a fan frequency converter. Background At present, a main fan manufacturer mainly judges whether a cooling system fails according to the inlet and outlet pressure difference of a cooling pump, and the method can reflect the running condition of the cooling system to a certain extent, but has obvious limitations. As the operational life of the fans increases, the cooling system and associated components may deteriorate to varying degrees, resulting in operating parameters that deviate from the design values. The conventional cooling system failure judging method is inaccurate and reliable in long-term operation, false alarms are easily caused, the number and the duration of unplanned shutdown of the fan are increased, and the operation efficiency and the economic benefit of the wind power plant are seriously affected. In order to reduce non-stop loss, part of stations adopt a mode of tampering sensor signals of a cooling system to shield alarm, and the action not only violates a related management system, but also brings serious potential safety hazards and increases the risk of production accidents. Therefore, the invention aims to provide a more accurate, reliable and robust cooling system failure judging method of a fan frequency converter, which can accurately judge the running state of the cooling system under the condition that the running period of a fan is increased, reduce false alarms, reduce the number of unplanned shutdown times, improve the running efficiency and the safety of a wind power generation system, avoid potential safety hazards caused by tampering sensor signals and ensure the stable and efficient running of the wind power generation system. Disclosure of Invention The invention aims to provide a cooling system failure judging method of a fan frequency converter, which effectively solves the technical problems in the background technology. In order to achieve the above purpose, the present invention provides the following technical solutions. A cooling system failure judging method of a fan frequency converter is characterized by comprising the following steps: S1, acquiring environmental parameters in a specific area within 6-12 months and all fan operation parameters of the same model, wherein the environmental parameters at least comprise parameters of environmental temperature, wind speed and power generation; s2, based on S1, acquiring historical data about environmental parameters and fan running states in the area in the period of 6-12 months, uniformly modeling all fans of the same model, and constructing a model base prediction model delta P_M_B of the pressure difference of all fan frequency converter cooling systems of the same model; S3, based on the S2, testing and calculating a model current prediction model delta P_M_C and a model basic prediction model delta P_M_B to obtain an attenuation coefficient eta; S4, acquiring environmental parameters in the last 6-12 months in the same area as the step S1 and all fan operation parameters of the same model, modeling each fan of the same model, and constructing a current-stage fan prediction model delta P_M_Ci of the pressure difference of all fan frequency converter cooling systems of the same model; S5, based on S4, comparing the current cooling system pressure difference value of the fan frequency converter with a preset safety threshold T, and constructing a safety threshold judgment model, wherein if the current cooling system pressure difference value of the fan frequency converter is within a safety range, the threshold judgment model result takes a value of 0, and if the current cooling system pressure difference value of the fan frequency converter is not within the safety range, the result takes a value of 1, and the threshold judgment model result is represented by a; S6, based on the S5, calculating according to the current cooling system pressure difference value of the fan frequency converter and the predicted value of the current fan prediction model delta P_M_Ci, and constructing an individual deviation judgment model; s7, based on the S5, calculating according to the current cooling system pressure difference value of the fan frequency converter and the predicted value of the current fan prediction model delta P_M_Ci, and constructing a model deviation judgment model; S8, based on S6 and S7, constructing a comprehensive judgment model f according to the attenuation coefficient eta, the threshold judgment model result, the individual deviation judgment model result and the model deviation judgment model result to obtain a comprehensive judgment coefficient Z, and triggering a cooling system failure alarm if the calculated Z value is larger than a set threshold T, otherwise, considering that the fan can still normall