CN-121998126-A - System and method for estimating power consumption of vehicle

CN121998126ACN 121998126 ACN121998126 ACN 121998126ACN-121998126-A

Abstract

The invention provides an estimation system for estimating power consumption of a vehicle which appropriately judges the necessity of learning and performs learning. A power consumption estimation system for a vehicle includes a1 st storage unit that stores a prediction error of power consumption of the vehicle, a 2 nd storage unit that stores an input/output pair of a model for estimating power consumption of the vehicle, and a learning unit that, when the estimation accuracy of the model for estimating power consumption of the vehicle is poor, uses an average and standard deviation of the prediction error of power consumption of the vehicle in the 1 st storage unit to determine whether improvement of estimation accuracy can be expected by learning using the 2 nd storage unit, and learns only when improvement of estimation accuracy can be expected.

Inventors

Song Banxiuwu
GUAN XIAOYI
Hizawa Yingming

Assignees

丰田自动车株式会社

Dates

Publication Date: 20260508
Application Date: 20251022
Priority Date: 20241107

Claims (5)

1. An estimation system for power consumption of a vehicle is characterized by comprising: A1 st storage unit that stores a prediction error of power consumption of the vehicle; A2 nd storage unit for storing an input/output pair of a model for estimating power consumption of the vehicle, and And a learning unit that, when an estimation accuracy of a model for estimating the power consumption of the vehicle is poor, uses an average and standard deviation of a prediction error of the power consumption of the vehicle in the 1 st storage unit to determine whether improvement of the estimation accuracy can be expected by learning using the 2 nd storage unit, and performs learning only when improvement of the estimation accuracy can be expected.
2. The system for estimating power consumption of a vehicle according to claim 1, wherein, When the average is small and the standard deviation is small, it is determined that improvement of the estimation accuracy cannot be expected, When the average is large and the standard deviation is small, it is determined that improvement of the estimation accuracy can be expected, When the average is small and the standard deviation is large, it is determined that improvement of the estimation accuracy cannot be expected.
3. The system for estimating power consumption of a vehicle according to claim 1, wherein, The 1 st storage unit and the 2 nd storage unit store data for each driver or each vehicle.
4. The system for estimating power consumption of a vehicle according to claim 1, wherein, And removing abnormal data of the power consumption of the vehicle to calculate the average and standard deviation of the prediction error of the power consumption.
5. A method for estimating power consumption of a vehicle is characterized by comprising: A1 st storage unit for storing a prediction error of the power consumption of the vehicle, and A2 nd storage unit that stores an input/output pair of a model for estimating the power consumption of the vehicle, In the method, when the estimation accuracy of the model for estimating the power consumption of the vehicle is poor, the average and standard deviation of the prediction error of the power consumption of the vehicle in the 1 st storage unit are used to determine whether the improvement of the estimation accuracy can be expected by the learning using the 2 nd storage unit, and the learning is performed only when the improvement of the estimation accuracy can be expected.

Description

System and method for estimating power consumption of vehicle Technical Field The present invention relates to a system and a method for estimating power consumption of a vehicle. Background Patent document 1 describes a device for obtaining appropriate fuel economy or electric energy efficiency in accordance with the running condition of a vehicle and a driver. Patent document 1 Japanese patent application laid-open No. 2017-9460 Disclosure of Invention However, the apparatus of patent document 1 is set to perform learning for each elapsed time, and thus performs learning at a time when learning is not necessary. Accordingly, an object of the present invention is to provide an estimation system of power consumption of a vehicle that appropriately determines the necessity of learning and performs learning. The power consumption estimation system for a vehicle according to the present invention includes: A1 st storage unit that stores a prediction error of power consumption of the vehicle; A2 nd storage unit for storing an input/output pair of a model for estimating power consumption of the vehicle, and And a learning unit that, when an estimation accuracy of a model for estimating the power consumption of the vehicle is poor, uses an average and standard deviation of a prediction error of the power consumption of the vehicle in the 1 st storage unit to determine whether improvement of the estimation accuracy can be expected by learning using the 2 nd storage unit, and performs learning only when improvement of the estimation accuracy can be expected. With the above configuration, an estimation system of the power consumption of the vehicle that appropriately judges the necessity of learning and performs learning is provided. Machine learning is used for learning. The power consumption estimation system of the vehicle of the invention is characterized in that, When the average is small and the standard deviation is small, it is determined that improvement of the estimation accuracy cannot be expected, When the average is large and the standard deviation is small, it is determined that improvement of the estimation accuracy can be expected, When the average is small and the standard deviation is large, it is determined that improvement of the estimation accuracy cannot be expected. The above-described configuration is an example of a determination method using the average and standard deviation of the prediction error. The power consumption estimation system of the vehicle of the invention is characterized in that, The 1 st storage unit and the 2 nd storage unit store data for each driver or each vehicle. With the above configuration, the variation in power consumption by the driver or the vehicle can be considered. The power consumption estimation system of the vehicle of the invention is characterized in that, And removing abnormal data of the power consumption of the vehicle to calculate the average and standard deviation of the prediction error of the power consumption. With the above configuration, an estimation system of the power consumption of the vehicle that further accurately and appropriately judges the necessity of learning and performs learning is provided. The method for estimating the power consumption of a vehicle according to the present invention includes: A1 st storage unit for storing a prediction error of the power consumption of the vehicle, and A2 nd storage unit that stores an input/output pair of a model for estimating the power consumption of the vehicle, In the method, when the estimation accuracy of the model for estimating the power consumption of the vehicle is poor, the average and standard deviation of the prediction error of the power consumption of the vehicle in the 1 st storage unit are used to determine whether the improvement of the estimation accuracy can be expected by the learning using the 2 nd storage unit, and the learning is performed only when the improvement of the estimation accuracy can be expected. With the above configuration, an estimation method of the power consumption of the vehicle that appropriately judges the necessity of learning and performs learning is provided. Effects of the invention The present invention provides an estimation system and the like for estimating the power consumption of a vehicle that appropriately determines the necessity of learning and performs learning. Drawings Fig. 1A is a flowchart 1 of a method for estimating power consumption of a vehicle according to an embodiment. Fig. 1B is a2 nd flowchart of a method of estimating the power consumption of the vehicle according to the embodiment. Fig. 1C is a 3 rd flowchart of a method of estimating the power consumption of the vehicle according to the embodiment. Fig. 2 is a flowchart (a) and a diagram (B) showing a prediction error in a case where the 1 st learning is not performed in the method for estimating the power consumption amount of the vehicle according to the embodiment.