Hotel dish-washing machine load adjustable capability evaluation and power capacity market compensation method

Abstract

The invention discloses a hotel dish-washing machine load adjustable capability evaluation and power capacity market compensation method. The method comprises the following steps: acquiring hotel dish-washing machine load data and related data; predicting the guest flow volume of the hotel by using a multi-modal artificial intelligence algorithm; constructing a hotel dish-washing machine step load model based on the passenger flow volume, and obtaining a dish-washing machine step load by using the hotel dish-washing machine step load model; and based on the dish-washing machine step load, constructing a hotel dish-washing machine load day-ahead adjustable capability evaluation and power capacity market compensation model, evaluating the day-ahead up-regulation and down-regulation adjustable capability of the hotel dish-washing machine load, and respectively performing up-regulation capacity market compensation and down-regulation capacity market compensation for the hotel. The method can improve the prediction precision of the hotel passenger flow, quantitatively evaluate the day-ahead adjustable capability of the hotel dish-washing machine load, and improve the enthusiasm of the hotel to participate in the power capacity market.

Description

Technical Field

[0001] The invention relates to the technical field of power system regulation and control, and in particular to a method for evaluating the load adjustable capacity of a hotel dishwasher and a power capacity market compensation method. Background Art

[0002] The load of hotel dishwashers is large. Effective evaluation of the load of hotel dishwashers is of great significance to improving the regulation capacity of the load side and supporting grid regulation. By giving hotels a certain amount of power capacity market compensation, the enthusiasm of hotels to participate in grid operation can be improved. However, the existing methods have the following shortcomings: 1) Insufficient passenger flow prediction. The existing methods do not comprehensively consider new factors such as weather, temperature, and holidays. The current passenger flow prediction models are often based on historical data and simple time series analysis, lacking comprehensive consideration of external factors such as weather conditions (such as sunny days, rainy days), temperature changes, and holidays (such as holidays, weekends); since these factors have a significant impact on passenger flow, ignoring them will lead to inaccurate predictions, thereby affecting the evaluation and regulation strategy of dishwasher load. 2) The correlation model between passenger flow and dishwasher load is not detailed enough. The existing methods usually simply assume that there is a linear relationship between passenger flow and dishwasher load, and do not fully explore the complex correlation between the two. In fact, the fluctuation of passenger flow may be affected by the interaction of multiple factors, such as catering demand and special activities in the same period, and these factors are not carefully reflected in the correlation model, resulting in inaccurate load assessment results. 3) The day-ahead adjustable capacity evaluation model is not detailed enough. The existing day-ahead adjustable capacity evaluation model often lacks detailed consideration of power capacity compensation. The fluctuation of dishwasher load directly affects power consumption. However, the current evaluation methods usually ignore how to effectively compensate for power capacity when the load fluctuates, which will affect the reliable operation of the dishwasher, especially during high-load periods, which may lead to insufficient power supply or equipment overload. 4) Lack of dynamic adjustment mechanism. Many methods lack the ability to dynamically adjust the dishwasher load. They mainly rely on static plans on the day-ahead and cannot effectively respond to changes in the actual situation on the day. For example, if the passenger flow increases sharply due to sudden weather changes, the fixed scheduling plan may not be able to adjust the load of the dishwasher in time, resulting in dishwasher overload or reduced cleaning efficiency. 5) Underutilization of multimodal data. Although some methods introduce factors such as weather and holidays, the comprehensive utilization of multimodal data and the application of deep learning methods are still not comprehensive enough, and the implicit correlation between data has not been fully explored. This limits the predictive ability of the model, especially in complex operating environments. Summary of the invention

[0003] In order to solve the above problems, the present invention proposes a hotel dishwasher load adjustable capacity evaluation and power capacity market compensation method that can enhance the enthusiasm of hotels to participate in the power capacity market.

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

[0005] The present invention is a method for evaluating the load adjustable capacity of a hotel dishwasher and compensating for the power capacity market, comprising the following operations:

[0006] Collect hotel dishwasher load data and related data, including weather information, holiday information, temperature information, historical passenger flow information, dishwasher load operation time information, dishwasher power, and power capacity market compensation coefficient;

[0007] Use multimodal AI algorithms to predict hotel customer traffic;

[0008] Construct a hotel dishwasher step load model based on passenger flow, and use the hotel dishwasher step load model to obtain the dishwasher step load;

[0009] Based on the step load of dishwashers, a model for evaluating the day-ahead adjustable capacity of hotel dishwasher load and compensating in the electricity capacity market is constructed to evaluate the day-ahead adjustable capacity of hotel dishwasher load, and compensate the hotel through the capacity increase market and capacity decrease market respectively.

[0010] A further improvement of the present invention is that the expression of the collected hotel dishwasher load data and related data is:

[0011]

[0012] X2(t)=[t za ,t zb ,t wa ,t wb ,t z1 ,t z2 ,t w1 ,t w2 ,P0(t)]

[0013] Among them, X1(t) is the non-hotel dishwasher load information on the tth day, w(t) is the weather information on the tth day, with a value of 1 for sunny days, 2 for cloudy days, 3 for rainy days, and 4 for snowy days; d(t) is the holiday information on the tth day, with a value of 1 for holidays, 2 for weekends, and 3 for weekdays; T(t) is the temperature information on the tth day; L(t) is the historical passenger flow information on the tth day; The capacity compensation coefficient is increased in the electricity capacity market; is the capacity reduction compensation coefficient given by the power capacity market; X2(t) is the hotel dishwasher operation information on the tth day, t za The earliest time to start washing dishes is at noon, t zb The dishwashing deadline is noon; wais the earliest time to start washing dishes in the evening, t wb Cut off time for dishwashing in the evening; z1 is the dishwasher startup time at noon, t z2 The dishwasher shuts down at noon; w1 The dishwasher startup time at night, t w2 is the dishwasher shutdown time at night; P0(t) is the load of a single dishwasher in the hotel.

[0014] A further improvement of the present invention is to use a multimodal artificial intelligence algorithm to predict the hotel's passenger flow, specifically including:

[0015] The feature fusion method is used to combine data of different modes to form a multimodal feature vector of the hotel, which is expressed as:

[0016] X' d (t)=[w'(t),d'(t),T'(t),L'(t)]

[0017] Among them, X' d (t) is the fused hotel multimodal feature vector, w'(t) is the weather information vector after one-hot encoding, d'(t) is the holiday information vector after one-hot encoding, T'(t) is the standardized temperature value, and L'(t) is the standardized historical passenger flow value;

[0018] The hotel multimodal feature vector is input into the multilayer perceptron to predict the hotel's passenger flow. The objective function of the prediction is to minimize the error between the predicted value and the true value. The expression of the objective function is:

[0019]

[0020] Among them, δ(t) is the mean square error of hotel passenger flow predicted by the multimodal artificial intelligence algorithm, N is the number of days, L”(t) is the predicted hotel passenger flow, and L(t) is the historical passenger flow information on the tth day;

[0021] The expression for hotel passenger flow prediction is:

[0022] L”(t)=f mt (w'(t),d'(t),T'(t),L'(t))

[0023] Among them, f mt () is a multimodal artificial intelligence algorithm prediction model.

[0024] A further improvement of the present invention is that the expression of the step load model of the hotel dishwasher is:

[0025]

[0026] Among them, P(t) is the step load of the hotel dishwasher, M1, M2, M3, and M4 are the number of started dishwashers, P0(t) is the basic operating load of the dishwasher, γ1, γ2, γ3, and γ4 are the hotel passenger flow thresholds, and L”(t) is the predicted hotel passenger flow.

[0027] The further improvement of the present invention is that the expression of the hotel dishwasher load adjustable capacity evaluation and the power capacity market compensation model is:

[0028]

[0029]

[0030]

[0031]

[0032] Among them, ΔP up (t) is the capacity of the hotel dishwasher load that can be adjusted upwards a few days ago, ΔP dn (t) is the capacity of the hotel dishwasher to reduce load in the previous day, P z (t) is the hotel’s noon dishwasher load, P w (t) is the hotel’s dishwasher load at night, t za is the earliest time to start washing dishes at noon, t zb The dishwashing deadline is noon, t wa is the earliest time to start washing dishes in the evening, t wb The end time for washing dishes in the evening, t z1 The dishwasher startup time is noon, t z2 The dishwasher shut-off time is at noon, t w1 The dishwasher startup time at night, t w2 The dishwasher shuts down at night. up Market capacity compensation for hotel dishwasher load increase, Y dn Market capacity compensation for hotel dishwasher load reduction, The capacity compensation coefficient is increased by the power capacity market. It is the capacity compensation coefficient reduced given by the electricity capacity market.

[0033] The beneficial effects of the present invention are as follows: the present invention improves the prediction accuracy of hotel passenger flow, refines the association between passenger flow and hotel dishwashers, quantifies the day-ahead adjustable capacity of evaluating hotel dishwasher loads, and through the hotel dishwasher load power capacity market compensation model, improves the enthusiasm of hotels to participate in the power capacity market and enhances the regulation capacity of the power grid. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 is a flow chart in an embodiment of the present invention;

[0035] Figure 2 Schematic diagram of load working characteristics of a hotel dishwasher in an embodiment of the present invention. DETAILED DESCRIPTION

[0036] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain 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 can be combined with each other as long as they do not conflict with each other.

[0037] like Figure 1 As shown, the hotel dishwasher load adjustable capacity evaluation and power capacity market compensation method of this embodiment includes the following steps:

[0038] Step 1: Collect the hotel dishwasher load data and related data, including weather information, holiday information, temperature information, historical passenger flow information, dishwasher load operation time information, dishwasher power, power capacity market compensation coefficient, etc. The dishwasher load operation time information includes the earliest time to start washing dishes at noon and evening, the latest time to end washing dishes, the dishwasher startup time, and the dishwasher shutdown time. The collected data is expressed as:

[0039]

[0040] X2(t)=[t za ,t zb ,t wa ,t wb ,t z1 ,t z2 ,t w1 ,t w2 ,P0(t)](2)

[0041] Among them, X1(t) is the non-hotel dishwasher load information on the tth day, w(t) is the weather information on the tth day, with a value of 1 for sunny days, 2 for cloudy days, 3 for rainy days, and 4 for snowy days; d(t) is the holiday information on the tth day, with a value of 1 for holidays, 2 for weekends, and 3 for weekdays; T(t) is the temperature information on the tth day; L(t) is the historical passenger flow information on the tth day, including the noon passenger flow and the evening passenger flow; The capacity compensation coefficient is increased in the electricity capacity market; is the capacity reduction compensation coefficient given by the power capacity market; X2(t) is the hotel dishwasher operation information on the tth day, t zaThe earliest time to start washing dishes is at noon, t zb The dishwashing deadline is noon; wa is the earliest time to start washing dishes in the evening, t wb Cut off time for dishwashing in the evening; z1 is the dishwasher startup time at noon, t z2 The dishwasher shuts down at noon; w1 The dishwasher startup time at night, t w2 is the dishwasher shutdown time at night; P0(t) is the load of a single dishwasher in the hotel.

[0042] Step 2: Use a multimodal artificial intelligence algorithm to predict the hotel's passenger flow. The multimodal artificial intelligence algorithm can train data of different modes to improve the effectiveness and accuracy of the prediction. This embodiment uses a multimodal artificial intelligence algorithm to predict the hotel's passenger flow. The multimodal artificial intelligence algorithm is used to train the weather information w(t) on the tth day, the holiday information d(t) on the tth day, the temperature information T(t) on the tth day, and the historical passenger flow information L(t) on the tth day to predict the hotel's passenger flow. Specifically, it includes:

[0043] First, the feature fusion method is used to combine the data of different modes to form a unified feature vector, namely the hotel multimodal feature vector, as shown in expression (3). This is a feature vector spliced ​​from the input of different modes, representing all the relevant input data of the hotel passenger flow at the current time t:

[0044] X' d (t)=[w'(t),d'(t),T'(t),L'(t)] (3)

[0045] Among them, X' d (t) is the fused hotel multimodal feature vector, w'(t) is the weather information vector after One-Hot Encoding, d'(t) is the holiday information vector after One-Hot Encoding, T'(t) is the standardized temperature value, and L'(t) is the standardized historical passenger flow value;

[0046] Secondly, the hotel multimodal feature vector X' d (t) is input into a multi-layer perceptron (MLP) and the hotel passenger flow is predicted through a series of fully connected layers. The objective function of the prediction in this embodiment is to minimize the error between the predicted value and the true value. In this embodiment, the mean square error (MSE) is minimized, as shown in expression (4):

[0047]

[0048] Among them, δ(t) is the mean square error of hotel passenger flow predicted by the multimodal artificial intelligence algorithm, N is the number of days, and L”(t) is the predicted hotel passenger flow.

[0049] After the above training, the hotel passenger flow prediction based on multimodal artificial intelligence can be obtained, as shown in expression (5), the input hotel multimodal feature vector X' d (t), we can predict the hotel’s passenger flow L”(t) for the next day.

[0050] L”(t)=f mt (w'(t),d'(t),T'(t),L'(t)) (5)

[0051] Among them, f mt () is a multimodal artificial intelligence algorithm prediction model.

[0052] Step 3, construct a hotel dishwasher step load model based on passenger flow. The hotel chooses to start different numbers of dishwashers according to the size of the passenger flow to ensure the completion of the dishwashing task. When the passenger flow is small, the number of dishwashers started is small, and when the passenger flow is large, multiple dishwashers need to be started at the same time. This embodiment uses a hotel dishwasher step load model based on passenger flow to accurately describe the hotel dishwasher process and the resulting dishwasher step load. The hotel dishwasher step load model based on passenger flow is shown in expression (6):

[0053]

[0054] Among them, P(t) is the step load of the hotel dishwasher, M1, M2, M3, and M4 are the number of started dishwashers, P0(t) is the basic operating load of the dishwasher, γ1, γ2, γ3, and γ4 are the hotel passenger flow thresholds, and L”(t) is the predicted hotel passenger flow.

[0055] Step 4, construct a hotel dishwasher load adjustable capacity assessment and power capacity market compensation model. Based on the constructed hotel dishwasher step load model, this embodiment constructs a hotel dishwasher load adjustable capacity assessment model to assess the hotel dishwasher load's ability to adjust upward and downward the day before, providing support for the dispatch of the new power system. Based on the assessed adjustable capacity, a hotel dishwasher load power capacity market compensation model is constructed to compensate the hotel with the upward and downward capacity markets, respectively, to increase the hotel's enthusiasm for participating in the power capacity market and enhance the power grid's regulation capacity.

[0056] Hotel dishwasher working characteristics Figure 2As shown, the hotel dishwasher load usually works at noon and evening, and usually starts washing dishes after lunch and dinner. During the dishwashing stage, it can participate in the grid regulation and provide the grid with upward and downward regulation capabilities by moving the load forward or backward. The constructed hotel dishwasher load day-ahead adjustable capacity evaluation model is shown in Expressions (7) and (8):

[0057]

[0058] Among them, ΔP up (t) is the capacity of the hotel dishwasher load that can be adjusted upwards a few days ago, ΔP dn (t) is the capacity of the hotel dishwasher to reduce load in the previous day, P z (t) is the hotel’s noon dishwasher load, P w (t) is the hotel's dishwasher load at night. z (t), P w (t) is calculated based on the hotel's noon and evening passenger flow predicted in step 2 and further calculated based on expression (6) in step 3.

[0059] The hotel dishwasher load electricity capacity market compensation model is shown in expressions (9) and (10), where expression (9) is the hotel dishwasher load increase market capacity compensation model, and expression (10) is the hotel dishwasher load decrease market capacity compensation model.

[0060]

[0061] Among them, Y up Market capacity compensation for hotel dishwasher load increase, Y dn Market capacity compensation for hotel dishwasher load reduction, The capacity compensation coefficient is increased by the power capacity market. It is the capacity compensation coefficient reduced given by the electricity capacity market.

[0062] Through the above steps, the day-ahead adjustable capacity of the hotel's dishwasher load can be effectively evaluated, and compensation can be given in the electricity capacity market, thereby increasing the enthusiasm of hotels to participate in the electricity capacity market and improving the regulation capacity of the power grid.

[0063] 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 those generally 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 the same meaning as in the context of the prior art, and will not be interpreted with an idealized or overly formal meaning unless defined similarly here.

[0064] 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. Hotel dishwasher load adjustable capacity assessment and power capacity market compensation method, characterized by: The following operations are included: Collect hotel dishwasher load data and related data, including weather information, holiday information, temperature information, historical passenger flow information, dishwasher load operation time information, dishwasher power, and power capacity market compensation coefficient; Use multimodal AI algorithms to predict hotel customer traffic; Construct a hotel dishwasher step load model based on passenger flow, and use the hotel dishwasher step load model to obtain the dishwasher step load; Based on the step load of dishwashers, a model for evaluating the day-ahead adjustable capacity of hotel dishwasher load and compensating in the electricity capacity market is constructed to evaluate the day-ahead adjustable capacity of hotel dishwasher load, and compensate the hotel through the capacity increase market and capacity decrease market respectively.

2. The method for evaluating the load adjustable capacity of a hotel dishwasher and compensating for power capacity in a market according to claim 1, characterized in that: The expression of the collected hotel dishwasher load data and related data is: X2(t)=[t za ,t zb ,t wa ,t wb ,t z1 ,t z2 ,t w1 ,t w2 ,P0(t)] Among them, X1(t) is the non-hotel dishwasher load information on the tth day, w(t) is the weather information on the tth day, with a value of 1 for sunny days, 2 for cloudy days, 3 for rainy days, and 4 for snowy days; d(t) is the holiday information on the tth day, with a value of 1 for holidays, 2 for weekends, and 3 for weekdays; T(t) is the temperature information on the tth day; L(t) is the historical passenger flow information on the tth day; The capacity compensation coefficient is increased in the electricity capacity market; is the capacity reduction compensation coefficient given by the power capacity market; X2(t) is the hotel dishwasher operation information on the tth day, t za The earliest time to start washing dishes is at noon, t zb The dishwashing deadline is noon; wa is the earliest time to start washing dishes in the evening, t wb Cut off time for dishwashing in the evening; z1 is the dishwasher startup time at noon, t z2 The dishwasher shuts down at noon; w1 The dishwasher startup time at night, t w2 is the dishwasher shutdown time at night; P0(t) is the load of a single dishwasher in the hotel.

3. The method for evaluating the load adjustable capacity of a hotel dishwasher and compensating for power capacity in a market according to claim 1, characterized in that: The method of using a multimodal artificial intelligence algorithm to predict the hotel's passenger flow specifically includes: The feature fusion method is used to combine data of different modes to form a multimodal feature vector of the hotel, which is expressed as: X' d (t)=[w'(t),d'(t),T'(t),L'(t)] Among them, X' d (t) is the fused hotel multimodal feature vector, w'(t) is the weather information vector after one-hot encoding, d'(t) is the holiday information vector after one-hot encoding, T'(t) is the standardized temperature value, and L'(t) is the standardized historical passenger flow value; The hotel multimodal feature vector is input into the multilayer perceptron to predict the hotel's passenger flow. The objective function of the prediction is to minimize the error between the predicted value and the true value. The expression of the objective function is: Among them, δ(t) is the mean square error of hotel passenger flow predicted by the multimodal artificial intelligence algorithm, N is the number of days, L”(t) is the predicted hotel passenger flow, and L(t) is the historical passenger flow information on the tth day; The expression for hotel passenger flow prediction is: L”(t)=f mt (w'(t),d'(t),T'(t),L'(t)) Among them, f mt () is a multimodal artificial intelligence algorithm prediction model.

4. The method for evaluating the load adjustable capacity of a hotel dishwasher and compensating for power capacity in a market according to claim 1, characterized in that: The expression of the hotel dishwasher step load model is: Among them, P(t) is the step load of the hotel dishwasher, M1, M2, M3, and M4 are the number of started dishwashers, P0(t) is the basic operating load of the dishwasher, γ1, γ2, γ3, and γ4 are the hotel passenger flow thresholds, and L”(t) is the predicted hotel passenger flow.

5. The method for evaluating the load adjustable capacity of a hotel dishwasher and compensating for power capacity in a market according to claim 1, characterized in that: The expression of the hotel dishwasher load day-ahead adjustable capacity evaluation and power capacity market compensation model is: Among them, ΔP up (t) is the capacity of the hotel dishwasher load that can be adjusted upwards a few days ago, ΔP dn (t) is the capacity of the hotel dishwasher to reduce load in the previous day, P z (t) is the hotel’s noon dishwasher load, P w (t) is the hotel’s dishwasher load at night, t za is the earliest time to start washing dishes at noon, t zb The dishwashing deadline is noon, t wa is the earliest time to start washing dishes in the evening, t wb The end time for washing dishes in the evening, t z1 The dishwasher startup time is noon, t z2 The dishwasher shut-off time is at noon, t w1 The dishwasher startup time at night, t w2 The dishwasher shuts down at night. up Market capacity compensation for hotel dishwasher load increase, Y dn Market capacity compensation for hotel dishwasher load reduction, The capacity compensation coefficient is increased by the power capacity market. It is the capacity compensation coefficient reduced given by the electricity capacity market.

Images