Method and system for evaluating maximum load and residual adjustable capability of washing machine based on LSTM (Long Short Term Memory)

Abstract

The invention discloses an LSTM-based washing machine load maximum and residual adjustable capability evaluation method and system. The method comprises the steps of collecting user washing machine load data; constructing a single washing machine user power consumption behavior prediction model based on the LSTM network, and predicting the washing machine power consumption behavior of the user by using the single washing machine user power consumption behavior prediction model based on the LSTM network; constructing a washing machine load maximum adjustable capability evaluation model, and evaluating the maximum adjustable capability and the maximum adjustable capability of the washing machine user by using the washing machine load maximum adjustable capability evaluation model; and constructing a washing machine load residual adjustable capability evaluation model, and evaluating the residual adjustable capability of the washing machine load participating in power grid dispatching again after the washing machine load participates in dispatching by using the washing machine load residual adjustable capability evaluation model. The method improves the prediction precision of the load power consumption behavior of the washing machine, achieves the refined and quantitative evaluation of the load adjustable capability of the washing machine, and improves the power grid adjustment capability.

Description

Technical Field

[0001] The present invention relates to the technical field of power system control, and in particular to a method and system for evaluating the maximum load and remaining adjustable capacity of a washing machine based on LSTM. Background Art

[0002] Washing machine load assessment is crucial for improving the regulation capacity of urban power grids. Washing machine operation can be flexibly scheduled, allowing users to use the machine during periods of lower electricity prices or grid load. This distributes peak loads and reduces pressure on the grid during peak hours, thereby achieving peak load shifting and valley shifting. As a controllable load, washing machines can participate in demand response programs. When grid load is excessive or power supply is tight, washing machine operation can be delayed through remote control or user guidance to reduce instantaneous load and improve grid stability. By rationally scheduling washing machine operation and optimizing energy use, energy efficiency can be improved, energy waste can be reduced, and ultimately energy conservation and emission reduction can be achieved. However, existing assessments of washing machine load adjustability suffer from the following deficiencies: 1) Inaccurate predictions of electricity consumption behavior. Predictions of users' washing machine usage habits and electricity consumption behavior are insufficiently precise, failing to account for the diversity and variability of user behavior, making it impossible to accurately assess the potential for load regulation. 2) Rough assessments of load adjustability fail to provide a detailed analysis of the actual load adjustability of washing machines. For example, the maximum adjustable capacity (upward and downward) under different time periods and user scenarios is not specifically assessed. 3) Insufficient dynamic adjustment capabilities. After participating in grid dispatch, the remaining adjustment capacity of washing machines is often not fully assessed, making it impossible to adjust and respond to grid demands in real time. This limits the flexibility and effectiveness of washing machines as load regulation resources. 4) Data and algorithm limitations. Existing load adjustability assessments may rely on traditional statistical methods and limited data, failing to fully utilize advanced technologies such as big data analysis and machine learning, resulting in inaccurate and unreliable assessment results. Summary of the Invention

[0003] In order to solve the above problems, the present invention proposes a LSTM-based washing machine load maximum and remaining adjustable capacity evaluation method and system, which can improve the accuracy of washing machine power consumption behavior prediction.

[0004] In order to achieve the above object, the present invention is implemented through the following technical solutions:

[0005] The LSTM-based method for evaluating the maximum load and remaining adjustable capacity of a washing machine of the present invention includes the following operations:

[0006] Collect user washing machine load data, including washing machine startup time, washing machine shutdown time, washing machine operating time range, washing machine power;

[0007] Based on the user's washing machine load data, a single washing machine user electricity consumption behavior prediction model based on the LSTM network is constructed, and the user's washing machine electricity consumption behavior prediction model based on the LSTM network is used to predict the user's washing machine electricity consumption behavior;

[0008] Based on the prediction results of the user's washing machine electricity consumption behavior, a washing machine load maximum adjustable capacity evaluation model is constructed, and the maximum adjustable capacity and maximum adjustable capacity of the washing machine user are evaluated using the washing machine load maximum adjustable capacity evaluation model;

[0009] A washing machine load residual adjustable capacity evaluation model is constructed, and the washing machine load residual adjustable capacity evaluation model is used to evaluate the residual adjustable capacity of the washing machine load after it participates in the grid dispatch and then participates in the grid dispatch again.

[0010] A further improvement of the present invention is that the expression of the collected user washing machine load data is:

[0011]

[0012] Among them, X i (k) is the basic data of washing machine load of user i on day k, is the washing machine startup time of user i on day k, is the shutdown time of user i’s washing machine on day k, is the operating time range of the washing machine of user i on day k, where is the earliest possible startup time for user i on day k, is the latest startup time for user i on day k, P i is the washing machine power of user i.

[0013] A further improvement of the present invention is that the expression of the power consumption behavior prediction model of a single washing machine user based on the LSTM network is:

[0014]

[0015] Among them, f yi () is the LSTM training model for washing machine users’ electricity consumption behavior, are the predicted start-up time and shutdown time of the washing machine user, and Δk is the prediction step size.

[0016] A further improvement of the present invention is that the expression of the evaluation model of the maximum adjustable load capacity of the washing machine is:

[0017]

[0018] Where P(t) is the sum of the washing machine loads at time t, P i is the washing machine power of user i, is the time when user i's washing machine is turned on, is the shutdown time of user i’s washing machine, and M is the total number of washing machines in operation at time t;

[0019] The expression of the maximum adjustable load capacity model of the washing machine is:

[0020]

[0021] Among them, P up (t) is the maximum load adjustment capacity of the washing machine at time t, is the washing machine power that can be increased for user i1, K is the total number of washing machine loads that can be moved forward, To increase the earliest possible startup time of user i1, To increase the user i1 boot time;

[0022] The expression of the maximum load reduction capability model of the washing machine is:

[0023]

[0024] Among them, P dn (t) is the maximum load reduction capacity of the washing machine at time t, is the washing machine power that can be lowered for user i2, J is the total number of washing machine loads that can be moved back, This allows you to lower the latest time that the user's i2 can be turned on.

[0025] A further improvement of the present invention is that: the washing machine load remaining adjustable capacity evaluation model includes a washing machine load remaining upward adjustable capacity evaluation model and a washing machine load remaining downward adjustable capacity evaluation model;

[0026] The expression of the evaluation model of the washing machine load remaining adjustable capacity is:

[0027]

[0028]

[0029] Where ΔP up is the sum of the washing machine loads that can be moved forward between t1 and t2, P i is the washing machine power of user i, is the time when user i's washing machine is turned on, The latest time that user i can start the computer. is the remaining adjustable capacity at time t2, ΔP up (t1) is the load amount that responds to the power grid increase at time t1, P up (t2) is the original adjustable capacity at time t2;

[0030] The expression of the evaluation model of the remaining downward adjustment capacity of the washing machine load is:

[0031]

[0032]

[0033] Where ΔP dn is the sum of the loads transferred to the washing machine between t1 and t2, is the shutdown time of user i’s washing machine, is the remaining down-regulation capacity at time t2, ΔP dn (t1) is the load that responds to the power grid down-regulation at time t1, P dn (t2) is the original downward adjustable capacity at time t2.

[0034] The LSTM-based washing machine maximum load and remaining adjustable capacity evaluation system of the present invention includes:

[0035] The collection module is used to collect the user's washing machine load data, including the washing machine startup time, washing machine shutdown time, washing machine operating time range, and washing machine power;

[0036] A prediction model building module is used to build a single washing machine user electricity consumption behavior prediction model based on an LSTM network, and use the single washing machine user electricity consumption behavior prediction model based on an LSTM network to predict the user's washing machine electricity consumption behavior;

[0037] A maximum adjustable capacity evaluation model construction module is used to construct a washing machine load maximum adjustable capacity evaluation model, and use the washing machine load maximum adjustable capacity evaluation model to evaluate the maximum upward and maximum downward adjustable capacity of the washing machine user;

[0038] The remaining adjustable capacity evaluation model construction module is used to construct a washing machine load remaining adjustable capacity evaluation model, and use the washing machine load remaining adjustable capacity evaluation model to evaluate the remaining adjustable capacity of the washing machine load after participating in the scheduling and then participating in the power grid scheduling again.

[0039] A further improvement of the present invention is that the expression of the user washing machine load data collected by the collection module is:

[0040]

[0041] Among them, X i (k) is the basic data of washing machine load of user i on day k, is the washing machine startup time of user i on day k, is the shutdown time of user i’s washing machine on day k, is the operating time range of the washing machine of user i on day k, where is the earliest possible startup time for user i on day k, is the latest startup time for user i on day k, P i is the washing machine power of user i.

[0042] A further improvement of the present invention is that the expression of the power consumption behavior prediction model of a single washing machine user based on the LSTM network is:

[0043]

[0044] Among them, f yi () is the LSTM training model for washing machine users’ electricity consumption behavior, are the predicted start-up time and shutdown time of the washing machine user, and Δk is the prediction step size.

[0045] A further improvement of the present invention is that the expression of the evaluation model of the maximum adjustable load capacity of the washing machine is:

[0046]

[0047] Where P(t) is the sum of the washing machine loads at time t, P i is the washing machine power of user i, is the time when user i's washing machine is turned on, is the shutdown time of user i’s washing machine, and M is the total number of washing machines in operation at time t;

[0048] The expression of the maximum adjustable load capacity model of the washing machine is:

[0049]

[0050] Among them, P up (t) is the maximum load adjustment capacity of the washing machine at time t, is the washing machine power that can be increased for user i1, K is the total number of washing machine loads that can be moved forward, To increase the earliest possible startup time of user i1, To increase the user i1 boot time;

[0051] The expression of the maximum load reduction capability model of the washing machine is:

[0052]

[0053] Among them, P dn (t) is the maximum load reduction capacity of the washing machine at time t, is the washing machine power that can be lowered for user i2, J is the total number of washing machine loads that can be moved back, This allows you to lower the latest time that the user's i2 can be turned on.

[0054] A further improvement of the present invention is that: the washing machine load remaining adjustable capacity evaluation model includes a washing machine load remaining upward adjustable capacity evaluation model and a washing machine load remaining downward adjustable capacity evaluation model;

[0055] The expression of the evaluation model of the washing machine load remaining adjustable capacity is:

[0056]

[0057]

[0058] Where ΔP up is the sum of the washing machine loads that can be moved forward between t1 and t2, P i is the washing machine power of user i, is the time when user i's washing machine is turned on, The latest time that user i can start the computer. is the remaining adjustable capacity at time t2, ΔP up (t1) is the load amount that responds to the power grid increase at time t1, P up (t2) is the original adjustable capacity at time t2;

[0059] The expression of the evaluation model of the remaining downward adjustment capacity of the washing machine load is:

[0060]

[0061]

[0062] Where ΔP dn is the sum of the loads transferred to the washing machine between t1 and t2, is the shutdown time of user i’s washing machine, is the remaining down-regulation capacity at time t2, ΔP dn (t1) is the load that responds to the power grid down-regulation at time t1, P dn (t2) is the original downward adjustable capacity at time t2.

[0063] The beneficial effects of the present invention are: improving the accuracy of washing machine power consumption behavior prediction based on the LSTM algorithm, constructing a maximum and remaining adjustable capacity evaluation model, realizing a refined and quantitative evaluation of the washing machine load adjustable capacity, and improving the regulation capacity of the power grid. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] Figure 1 is a flow chart of a method in an embodiment of the present invention;

[0065] Figure 2is a schematic diagram of collected washing machine load data in an embodiment of the present invention;

[0066] Figure 3 Schematic diagram of LSTM network training in an embodiment of the present invention;

[0067] Figure 4 is a schematic diagram of the prediction results of the user's washing machine electricity consumption behavior in an embodiment of the present invention;

[0068] Figure 5 is a schematic diagram of the load-adjustable capability of a washing machine in an embodiment of the present invention;

[0069] Figure 6 4 is a schematic diagram of the remaining adjustable capacity evaluation result in an embodiment of the present invention. DETAILED DESCRIPTION

[0070] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only intended to illustrate the present invention and are not intended to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below may be combined with each other as long as they do not conflict with each other.

[0071] like Figure 1 As shown, the LSTM-based method for evaluating the maximum load and remaining adjustable capacity of a washing machine in this embodiment includes the following operations:

[0072] Step 1: Collect the user's washing machine load data, including the washing machine startup time, washing machine shutdown time, washing machine operating time range, washing machine power and other data, to provide a basis for evaluating the washing machine load adjustable capacity.

[0073] Step 2: Build a prediction model for the electricity usage behavior of a single washing machine user based on an LSTM network. Since users' laundry activities are somewhat random, this embodiment uses user washing machine load data samples collected through LSTM network training to predict the electricity usage behavior of a single user's washing machine, understand the user's electricity usage characteristics, and provide support for evaluating the adjustable capacity of the washing machine load.

[0074] Step 3: Based on the predicted results of the user's washing machine electricity usage behavior, a washing machine load maximum adjustable capacity evaluation model is constructed. The washing machine load maximum adjustable capacity evaluation model includes a washing machine load maximum adjustable upward capacity model and a washing machine load maximum adjustable downward capacity model, which respectively evaluate the maximum adjustable upward capacity and maximum adjustable downward capacity of the washing machine user.

[0075] Step 4: Build a washing machine load remaining adjustable capacity assessment model to evaluate the remaining adjustable capacity of the washing machine load after it participates in grid scheduling, providing support for continuous grid scheduling. The washing machine load remaining adjustable capacity assessment model includes a washing machine load remaining adjustable capacity assessment model and a washing machine load remaining adjustable capacity assessment model.

[0076] In step 1, the collected user washing machine load data is as follows Figure 2 As shown, its expression is:

[0077]

[0078] Among them, X i (k) is the basic data of washing machine load of user i on day k, is the washing machine startup time of user i on day k, is the shutdown time of user i’s washing machine on day k, is the operating time range of the washing machine of user i on day k, where is the earliest possible startup time for user i on day k, is the latest startup time for user i on day k, P i is the washing machine power of user i.

[0079] In step 2, the LSTM network training process is as follows Figure 3 In this embodiment, the input sample is: The output is: Where Δk is the prediction step size. The training process is expressed as:

[0080]

[0081] Among them, σ is the activation function, f k is the output of the forget gate, W f 、b f is the corresponding forget gate matrix, i k is the output of the input gate, W j 、b j is the corresponding input gate weight matrix, C k-1 is the old cell state information, Add candidate status information for selection, C k For updated cell information, W C 、b C is the corresponding neuron matrix, o k is the output of the output gate, W o 、b o is the corresponding output gate matrix, h k is the output result, h k-1 is the hidden layer input.

[0082] The final expression of the LSTM network-based single washing machine user electricity consumption behavior prediction model is:

[0083]

[0084] Among them, f yi () is the LSTM training model for washing machine users’ electricity consumption behavior, The power consumption behavior prediction model of a single washing machine user based on LSTM network is used to predict the power consumption behavior of each washing machine user. The prediction results are as follows: Figure 4 shown.

[0085] In step 3, based on the electricity consumption behavior of individual washing machine users predicted by the electricity consumption behavior prediction model, a washing machine load maximum adjustable capacity evaluation model is constructed to evaluate the load of the washing machine user at any time t. The expression of the washing machine load maximum adjustable capacity evaluation model is:

[0086]

[0087] Where P(t) is the sum of the washing machine loads at time t, P i is the washing machine power of user i, is the time when user i's washing machine is turned on, is the shutdown time of user i’s washing machine, and M is the total number of washing machines in operation at time t.

[0088] A model for the maximum adjustable load capacity of washing machines is constructed, indicating that washing machines with a startup time of time t and an earliest startup time before time t can be scheduled in advance. That is, the startup time of the washing machine is time t. During the period between the startup time and the earliest startup time, the washing machine is scheduled in advance. The sum of its loads is the maximum adjustable capacity, which is expressed as:

[0089] The expression of the maximum adjustable load capacity model of the washing machine is:

[0090]

[0091] Among them, P up (t) is the maximum load adjustment capacity of the washing machine at time t, is the washing machine power that can be increased for user i1, K is the total number of washing machine loads that can be moved forward, To increase the earliest possible startup time of user i1, To increase the user i1 boot time;

[0092] The expression of the maximum load reduction capability model of the washing machine is:

[0093]

[0094] Among them, P dn (t) is the maximum load reduction capacity of the washing machine at time t, is the washing machine power that can be lowered for user i2, J is the total number of washing machine loads that can be moved back, This allows you to lower the latest time that the user's i2 can be turned on.

[0095] like Figure 5 As shown, the washing machine load remaining adjustable capacity evaluation model of this embodiment is used to evaluate the remaining adjustable capacity of the washing machine load at time t2 after the washing machine load participates in the grid dispatch at time t1. The washing machine load remaining adjustable capacity evaluation model is shown in equations (7) and (8). Equation (7) represents the sum of the washing machine loads that can be moved forward between t1 and t2. In equation (8), the load amount ΔP that responds to the grid increase at time t1 is up (t1) is divided into two cases: 1) When the load adjusted by the response grid is less than the sum of the washing machine loads that can be moved forward between t1 and t2, that is, ΔP up (t1)≤ΔP up When the remaining adjustable capacity at time t2 is not affected, it is still P up (t2). 2) When the load adjusted by the response grid is greater than the sum of the washing machine loads that can be moved forward between t1 and t2, that is, ΔP up (t1)>ΔP up When , it means that the remaining adjustable load needs to be dispatched after time t2. At this time, the remaining adjustable capacity at time t2 will be reduced by (ΔP up (t1)-ΔP up ), the remaining adjustable capacity at time t2 changes.

[0096]

[0097] Where ΔP up is the sum of the washing machine loads that can be moved forward between t1 and t2, P i is the washing machine power of user i, is the time when user i's washing machine is turned on, The latest time that user i can start the computer. is the remaining adjustable capacity at time t2, ΔP up (t1) is the load amount that responds to the power grid increase at time t1, P up (t2) is the original adjustable capacity at time t2.

[0098] The evaluation model of the remaining load reduction capacity of washing machines is shown in Equations (9) and (10). Equation (9) represents the sum of the washing machine loads that can be shifted back to between t1 and t2 at time t. In Equation (10), the load amount ΔP that responds to the grid reduction at time t1 is dn (t1) is divided into two cases: 1) When the load reduced by the responding grid is less than the sum of the washing machine loads transferred between t1 and t2, that is, ΔP dn (t1)≤ΔP dn When the remaining down-regulation capacity at time t2 is not affected, it is still P dn (t2). 2) When the load reduced by the responding grid is greater than the sum of the washing machine loads transferred between t1 and t2, that is, ΔP dn (t1)>ΔP dn When t2 is reached, it means that the load of the washing machine is increased. The load can be reduced, that is, the load capacity can be further increased by (ΔP dn (t1)-ΔP dn ), the remaining downward adjustment capacity at time t2 changes.

[0099]

[0100] Where ΔP dn is the sum of the loads transferred to the washing machine between t1 and t2, is the shutdown time of user i’s washing machine, is the remaining down-regulation capacity at time t2, ΔP dn (t1) is the load that responds to the power grid down-regulation at time t1, P dn (t2) is the original downward adjustable capacity at time t2.

[0101] The above method is implemented by the LSTM-based washing machine maximum load and remaining adjustable capacity evaluation system of this embodiment, which includes:

[0102] The collection module is used to collect the user's washing machine load data, including the washing machine startup time, washing machine shutdown time, washing machine operating time range, and washing machine power;

[0103] A prediction model building module is used to build a single washing machine user electricity consumption behavior prediction model based on an LSTM network, and use the single washing machine user electricity consumption behavior prediction model based on an LSTM network to predict the user's washing machine electricity consumption behavior;

[0104] A maximum adjustable capacity evaluation model construction module is used to construct a washing machine load maximum adjustable capacity evaluation model, and use the washing machine load maximum adjustable capacity evaluation model to evaluate the maximum upward and maximum downward adjustable capacity of the washing machine user;

[0105] The remaining adjustable capacity evaluation model construction module is used to construct a washing machine load remaining adjustable capacity evaluation model, and use the washing machine load remaining adjustable capacity evaluation model to evaluate the remaining adjustable capacity of the washing machine load after participating in the scheduling and then participating in the power grid scheduling again.

[0106] It will be understood by those skilled in the art that, unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by those skilled in the art in the art to which the present invention belongs. It should also be understood that terms such as those defined in common dictionaries should be understood to have meanings consistent with their meanings in the context of the prior art and, unless defined similarly as herein, will not be interpreted in an idealized or overly formal sense.

[0107] The specific implementation methods described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation method of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for evaluating the maximum load and remaining adjustable capacity of a washing machine based on LSTM, characterized by: The following operations are included: Collect user washing machine load data, including washing machine startup time, washing machine shutdown time, washing machine operating time range, and washing machine power; Based on the user's washing machine load data, a single washing machine user electricity consumption behavior prediction model based on LSTM network is constructed, and the user's washing machine electricity consumption behavior is predicted using the single washing machine user electricity consumption behavior prediction model based on LSTM network; According to the prediction results of the user's washing machine electricity consumption behavior, a washing machine load maximum adjustable capacity evaluation model is constructed, and the washing machine load maximum adjustable capacity evaluation model is used to evaluate the maximum adjustable capacity and maximum adjustable capacity of the washing machine user; A washing machine load residual adjustable capacity evaluation model is constructed, and the residual adjustable capacity of the washing machine load after participating in the dispatch and participating in the grid dispatch again is evaluated by using the washing machine load residual adjustable capacity evaluation model.

2. The LSTM-based washing machine load maximum and remaining adjustable capacity evaluation method according to claim 1, characterized in that: The expression of the collected user washing machine load data is: Among them, X i (k) is the basic washing machine load data of user i on day k, is the time when user i starts the washing machine on day k, is the shutdown time of the washing machine of user i on the kth day, is the operating time range of the washing machine of user i on day k, where is the earliest possible startup time for user i on day k, is the latest available startup time for user i on day k, P i is the washing machine power of user i.

3. The LSTM-based washing machine load maximum and remaining adjustable capacity evaluation method according to claim 2, characterized in that: The expression of the power consumption behavior prediction model of a single washing machine user based on the LSTM network is: Among them, f yi () is the LSTM training model for the electricity consumption behavior of washing machine users, are the predicted start-up time and shutdown time of the washing machine user, and Δk is the prediction step size.

4. The LSTM-based washing machine load maximum and remaining adjustable capacity evaluation method according to claim 2, characterized in that: The expression of the evaluation model of the maximum adjustable load capacity of the washing machine is: Where P(t) is the sum of the washing machine loads at time t, P i is the washing machine power of user i, is the time when the washing machine of user i is turned on, is the shutdown time of user i’s washing machine, and M is the total number of washing machines in operation at time t; The expression of the maximum adjustable capacity model of washing machine load is: Among them, P up (t) is the maximum load adjustment capacity of the washing machine at time t, is the washing machine power that can be adjusted upward for user i1, K is the total number of washing machine loads that can be moved forward, To adjust the earliest time that user i1 can start, To increase the user i1 power-on time; The expression of the maximum adjustable load model of the washing machine is: Among them, P dn (t) is the maximum load reduction capacity of the washing machine at time t, P i2 is the washing machine power that can be reduced for user i2, J is the total number of washing machine loads that can be moved back, The latest boot time of user i2 can be adjusted downward.

5. The LSTM-based washing machine load maximum and remaining adjustable capacity evaluation method according to claim 2, characterized in that: The washing machine load remaining adjustable capacity assessment model includes a washing machine load remaining adjustable upward capacity assessment model and a washing machine load remaining adjustable downward capacity assessment model; The expression of the evaluation model of the remaining adjustable capacity of washing machine load is: Among them, ΔP up is the sum of the washing machine loads that can be moved forward between t1 and t2, P i is the washing machine power of user i, t i p is the time when the washing machine of user i is turned on, The latest time that user i can start the computer. is the remaining adjustable capacity at time t2, ΔP up (t1) is the load amount that responds to the increase in the power grid at time t1, P up (t2) is the original adjustable capacity at time t2; The expression of the evaluation model of the remaining adjustable down capacity of the washing machine load is: Among them, ΔP dn is the sum of the washing machine loads transferred between t1 and t2, is the shutdown time of user i’s washing machine, is the remaining down-regulation capacity at time t2, ΔP dn (t1) is the load amount that responds to the grid down-regulation at time t1, P dn (t2) is the original downward adjustable capacity at time t2.

6. The LSTM-based washing machine load maximum and remaining adjustable capacity evaluation system is characterized by: The system comprises: The collection module is used to collect the user's washing machine load data, including the washing machine startup time, the washing machine shutdown time, the washing machine can run time range, and the washing machine power; A prediction model building module is used to build a prediction model for the electricity consumption behavior of a single washing machine user based on an LSTM network, and use the prediction model for the electricity consumption behavior of a single washing machine user based on an LSTM network to predict the user's washing machine electricity consumption behavior; A maximum adjustable capacity evaluation model building module is used to build a washing machine load maximum adjustable capacity evaluation model, and use the washing machine load maximum adjustable capacity evaluation model to evaluate the maximum upward adjustable capacity and maximum downward adjustable capacity of the washing machine user; The remaining adjustable capacity evaluation model construction module is used to construct a washing machine load remaining adjustable capacity evaluation model, and use the washing machine load remaining adjustable capacity evaluation model to evaluate the remaining adjustable capacity of the washing machine load after participating in the dispatch and participating in the power grid dispatch again.

7. The LSTM-based washing machine load maximum and remaining adjustable capacity evaluation system according to claim 6, characterized in that: The expression of the user washing machine load data collected by the collection module is: Among them, X i (k) is the basic washing machine load data of user i on day k, is the time when user i starts the washing machine on day k, is the shutdown time of the washing machine of user i on the kth day, is the operating time range of the washing machine of user i on day k, where is the earliest possible startup time for user i on day k, is the latest available startup time for user i on day k, P i is the washing machine power of user i.

8. The LSTM-based washing machine load maximum and remaining adjustable capacity evaluation system according to claim 7, characterized in that: The expression of the power consumption behavior prediction model of a single washing machine user based on the LSTM network is: Among them, f yi () is the LSTM training model for the electricity consumption behavior of washing machine users, are the predicted start-up time and shutdown time of the washing machine user, and Δk is the prediction step size.

9. The LSTM-based washing machine load maximum and remaining adjustable capacity evaluation system according to claim 7, characterized in that: The expression of the evaluation model of the maximum adjustable load capacity of the washing machine is: Where P(t) is the sum of the washing machine loads at time t, P i is the washing machine power of user i, is the time when the washing machine of user i is turned on, is the shutdown time of user i’s washing machine, and M is the total number of washing machines in operation at time t; The expression of the maximum adjustable capacity model of washing machine load is: Among them, P up (t) is the maximum load adjustment capacity of the washing machine at time t, P i1 is the washing machine power that can be adjusted upward for user i1, K is the total number of washing machine loads that can be moved forward, To adjust the earliest time that user i1 can start, To increase the user i1 power-on time; The expression of the maximum adjustable load model of the washing machine is: Among them, P dn (t) is the maximum load reduction capacity of the washing machine at time t, is the washing machine power that can be reduced for user i2, J is the total number of washing machine loads that can be moved back, The latest boot time of user i2 can be adjusted downward.

10. The LSTM-based washing machine load maximum and remaining adjustable capacity evaluation system according to claim 7, characterized in that: The washing machine load remaining adjustable capacity assessment model includes a washing machine load remaining adjustable upward capacity assessment model and a washing machine load remaining adjustable downward capacity assessment model; The expression of the evaluation model of the remaining adjustable capacity of washing machine load is: Among them, ΔP up is the sum of the washing machine loads that can be moved forward between t1 and t2, P i is the washing machine power of user i, is the time when the washing machine of user i is turned on, The latest time that user i can start the computer. is the remaining adjustable capacity at time t2, ΔP up (t1) is the load amount that responds to the increase in the power grid at time t1, P up (t2) is the original adjustable capacity at time t2; The expression of the evaluation model of the remaining adjustable down capacity of the washing machine load is: Among them, ΔP dn is the sum of the washing machine loads transferred between t1 and t2, is the shutdown time of user i’s washing machine, is the remaining down-regulation capacity at time t2, ΔP dn (t1) is the load amount that responds to the grid down-regulation at time t1, P dn (t2) is the original downward adjustable capacity at time t2.

Images