ANN-based waterworks online regulation capability evaluation method

Abstract

The invention discloses an ANN-based waterworks online regulation capability evaluation method, which comprises the following steps: firstly, providing an ANN-based water pump online regulation capability evaluation model, dynamically estimating the current water pump efficiency by using an artificial neural network, and calculating the real-time up-regulation or down-regulation power range of a water pump; then, an ANN-based blower online regulation capability model is provided, the real-time efficiency of a blower motor is dynamically predicted, and the up-regulation space and the down-regulation space of the power of the equipment in the current state are accurately quantified; secondly, online adjustment capability evaluation models of the dosing equipment, the sludge dewatering equipment and the lighting equipment are provided respectively, and the online adjustment capability of each equipment is calculated in real time; and finally, comprehensively evaluating the overall up-regulation / down-regulation capability of the waterworks through the waterworks on-line regulation capability evaluation model, and responding to the on-line regulation and control requirements of the power grid. According to the method, the adjustable potential of each key electric equipment of the waterworks can be accurately depicted, and the real-time performance and the accuracy of adjustment capability identification are improved.

Description

Technical Field

[0001] The present invention relates to the field of power systems, and in particular to an ANN-based online regulation capability evaluation method for a waterworks. Background Art

[0002] Evaluating a water plant's online regulation capabilities helps accurately understand its potential for adjusting power consumption under different operating conditions, providing the power grid with stable and flexible load regulation resources. Real-time evaluation effectively supports load-side response, improves the grid's peak-shaving capabilities, reduces system operating costs, and ensures power supply security and the achievement of energy conservation and emission reduction goals.

[0003] In the existing technology, the water plant regulation capacity assessment methods have the following deficiencies: 1) Lack of detailed modeling. Current methods mostly take the overall load as the object, and are not subdivided into specific equipment (such as water pumps, blowers, etc.), making it difficult to accurately assess the regulation potential of each key equipment. 2) Ignoring nonlinear characteristics. Traditional assessment methods are mostly based on linear regression or empirical formulas. It is difficult to effectively capture the complex nonlinear relationship between equipment operation and environmental variables, affecting the assessment accuracy. 3) Poor real-time and adaptability. Most methods rely on static parameters or historical means. It is difficult to respond to dynamic changes in operating status and external conditions, and the regulation capacity assessment is delayed. 4) Insufficient consideration of coupling between equipment. The operating logic and process constraints between various equipment in the water plant are not systematically considered, which can easily lead to distortion in the regulation capacity assessment.

[0004] Therefore, the evaluation of the regulation capacity of water plants needs further research to accurately characterize the adjustable potential of key electrical equipment in water plants, improve the real-time and accuracy of regulation capacity identification, support the formulation of load-side response strategies for power grids, and provide a scientific basis for the coordinated scheduling of flexible resources. Summary of the Invention

[0005] The problem to be solved by the present invention is to provide an ANN-based online regulation capacity evaluation method for a water plant, which is used to accurately describe the adjustable potential of each key electrical equipment in the water plant and improve the real-time and accuracy of regulation capacity identification.

[0006] The present invention adopts the following technical solution: a method for evaluating the online regulation capacity of a water plant based on ANN, comprising the following steps:

[0007] Step 1: Construct an ANN-based water pump online adjustment capability assessment model: Using an artificial neural network (ANN), input real-time flow, head, voltage, current, and temperature parameters to dynamically predict the current water pump efficiency. Combining the difference between the rated operating conditions and the current operating conditions, the power range that the water pump can adjust up or down in real time is calculated to evaluate the water pump's online adjustment potential.

[0008] Step 2: Build an ANN-based online blower adjustment capability model: Using the blower's real-time speed, load torque, current, voltage, and ambient temperature as input, the artificial neural network dynamically predicts the current blower motor efficiency. By comparing the difference between the predicted results and the rated load conditions, the power adjustment space of the blower in the current state is quantified, and the blower's adjustment capability is evaluated.

[0009] Step 3: Build an online adjustment capability assessment model for dosing equipment: Identify the on / off status of the dosing pump in real time, evaluate the full load as the downward adjustment space when it is on, and use the rated load as the upward adjustment space when it is off. Determine the relationship between the real-time on / off status and the rated power of the dosing equipment, and dynamically evaluate the online power adjustment potential of the dosing equipment.

[0010] Step 4. Construct an online adjustment capacity evaluation model for the sludge dewatering equipment: Based on the real-time operating status of the sludge dewatering equipment (operating or shut down), dynamically determine the current power adjustment space of the sludge dewatering equipment. When the equipment is in operation, the current power is used as the downward adjustment potential. When the equipment is shut down, the rated power is used as the upward adjustment potential. The real-time online adjustment capacity of the sludge dewatering equipment is estimated.

[0011] Step 5: Build a lighting equipment online adjustment capability assessment model: Determine the power adjustment range based on the real-time on / off status of each lighting fixture. Leverage real-time monitoring of lighting equipment operating data to determine whether the fixtures have on / off or power adjustment capabilities. This model also summarizes the overall load adjustment potential for refined power control within the factory.

[0012] Step 6. Construct an online regulation capacity assessment model for the waterworks: Build an online regulation capacity assessment model for integrated water pumps, blowers, dosing equipment, sludge dewatering equipment, and lighting equipment to assess the overall regulation up / down capability of the waterworks and respond to the online regulation needs of the power grid.

[0013] Preferably, in step 1, an artificial neural network is used to predict the pump efficiency in real time, and the formula for the collected information is as follows:

[0014] X=[x1,x2,x3,x4,x5] T

[0015] Where x1 is the pump flow rate, x2 is the pump head, x3 is the input power, x4 is the pump speed, x5 is the pump temperature, and the superscript T indicates transposition.

[0016] The water pump online regulation capacity evaluation model based on ANN has L hidden layers and the output h of the lth layer is (l) The formula is as follows:

[0017] h (l) =f (l) (W (l) h (l-1) +b(l) )l=1,2,…,L

[0018] Where h (l) is the output of layer l, W (l) is the weight matrix of the lth layer, b (l) is the bias term of the lth layer, f (l) is a nonlinear activation function;

[0019] The output layer is the efficiency prediction value. The mechanical efficiency prediction formula of the water pump is as follows:

[0020]

[0021] Where η p is the predicted value of the mechanical efficiency of the water pump, W p is the output layer weight vector, b p is the output layer bias term, and σ(·) is a function of the output range [0,1].

[0022] Preferably, the ANN-based water pump online regulation capacity evaluation model is trained using historical operation data to minimize the prediction error; the mean square error is used as the loss function to calculate the water pump efficiency loss, and the parameters are optimized by the gradient descent method to obtain a trained water pump online regulation capacity evaluation model.

[0023] Preferably, the real-time efficiency of the water pump is calculated using a trained water pump online regulation capability evaluation model, and the formula is as follows:

[0024] η p (t) = ANN p (x1(t),x2(t),x3(t),x4(t),x5(t))

[0025] Where: η p For the real-time efficiency of the water pump, ANN p is the trained neural network model, x1 is the pump flow rate, x2 is the pump head, x3 is the input power, x4 is the pump speed, and x5 is the pump temperature;

[0026] The adjustment capability of the pump's real-time load is determined based on the current operating flow and the allowable adjustment range. The formula is as follows:

[0027]

[0028] Where: The pump's upward adjustment capacity, is the pump's down-regulation capacity, Q(t) is the pump's current operating flow rate, and Q max is the maximum allowable flow rate, Q min is the minimum stable flow rate, ρ is the density of water, g is the acceleration due to gravity, ηm is the motor efficiency, η p is the mechanical efficiency of the water pump, and H is the water pump head.

[0029] Preferably, in the blower online adjustment capability model described in step 2, the real-time efficiency of the motor is calculated using the following formula:

[0030] η m (t) = ANN m (y1(t),y2(t),y3(t),y4(t),y5(t))

[0031] Where: η m Real-time efficiency of the motor, ANN m is the trained neural network model, y1 is the real-time speed of the blower, y2 is the load torque of the blower, y3 is the current of the blower, y4 is the voltage of the blower, and y5 is the ambient temperature;

[0032] Determine the online adjustment capability of the blower equipment based on the current load and the allowable variation range. The calculation is as follows:

[0033]

[0034] Where: For the blower's upward adjustment capability, is the blower's down-regulation capability, T(t) is the current load torque, T max is the maximum torque allowed by the load, T min is the minimum torque allowed by the load, η m is the motor efficiency and n is the motor speed.

[0035] Preferably, in the online adjustment capability model of the dosing equipment described in step 3, the dosing equipment operates intermittently and has on / off adjustment capability; when the dosing equipment is turned on, it has the capability to adjust all loads downward; when the dosing equipment is in the off state, it has the capability to adjust all loads upward; the upward and downward adjustment capabilities of the dosing equipment are calculated as follows:

[0036]

[0037] Where: To increase the capacity of the dosing equipment, For the down-regulation capacity of the dosing equipment, X d (t) is the operating status of the dosing equipment, P d,max The maximum rated power of the dosing equipment.

[0038] Preferably, in the online regulation capacity model of the sludge equipment in step 4, the sludge equipment is discretely regulated, with downward regulation capacity when running and upward regulation capacity when stopped; the upward regulation capacity and downward regulation capacity of the sludge equipment are calculated as follows:

[0039]

[0040] in: To increase the capacity of sludge equipment, For the down-regulation capacity of sludge equipment, X s (t) is the operating status of the sludge equipment, P s,max The maximum power rating of the device.

[0041] Preferably, in the lighting equipment online adjustment capability model described in step 5, the up / down adjustment capability of the lighting equipment is calculated as follows:

[0042]

[0043] Where: For lighting equipment up / down ability, X li (t) is whether the i-th lamp is adjustable, P unit,i is the rated power of the i-th lamp, n l is the number of lighting devices.

[0044] Preferably, the online regulation capability model of the water plant in step 6 and the specific steps of online regulation capability evaluation are as follows:

[0045] Step 6.1: Real-time collection of operating data of various equipment in the water plant, including flow rate, power, and speed; and real-time calculation of the online adjustment capacity of various equipment based on the rated parameters of the equipment;

[0046] Step 6.2: Summarize the online regulation capabilities of various equipment to obtain the overall online regulation capability of the water plant;

[0047] Step 6.3: Evaluate the real-time trend of online regulation capability to provide a basis for load regulation decisions.

[0048] Compared with the prior art, the present invention adopts the above technical solution and has the following technical effects:

[0049] 1. The online regulation capacity assessment method of the water plant of the present invention can accurately characterize the adjustable potential of each key electrical equipment in the water plant, improve the real-time and accuracy of regulation capacity identification, and support the formulation of load-side response strategies for the power grid.

[0050] 2. The overall regulation capacity model of the water plant constructed by the present invention can comprehensively evaluate the online regulation capacity of water pumps, blowers, dosing equipment, sludge dewatering equipment, and lighting equipment, provide a scientific basis for the coordinated scheduling of flexible resources, and realize refined and intelligent load management. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 This is a flowchart of the ANN-based water plant online regulation capacity evaluation method of the present invention. DETAILED DESCRIPTION

[0052] In order to make the purpose, technical solutions and advantages of the present invention clearer, the technical solutions of the application are further elaborated in detail below with reference to the accompanying drawings. The described embodiments are only a part of the embodiments involved in the present invention. All non-innovative embodiments of other researchers in this field on this embodiment fall within the scope of protection of the present invention. At the same time, the step numbers in the embodiments of the present invention are only set for the convenience of explanation and description, and the order between the steps is not limited in any way. The execution order of each step in the embodiment can be adaptively adjusted according to the understanding of those skilled in the art.

[0053] In one embodiment of the present invention, a method for evaluating the online regulation capability of a water plant based on ANN is proposed. Figure 1 As shown, it mainly includes: ANN-based water pump online adjustment capacity evaluation model, ANN-based blower online adjustment capacity evaluation model, dosing equipment online adjustment capacity evaluation model, sludge dewatering equipment online adjustment capacity evaluation model, lighting equipment online adjustment capacity evaluation model, and water plant online adjustment capacity evaluation model.

[0054] The online regulation capability evaluation of a waterworks using an ANN is performed, and the method includes the following steps:

[0055] Step 1: Construct an ANN-based water pump online regulation capability evaluation model.

[0056] Artificial neural networks are used to predict pump efficiency in real time. By inputting parameters such as real-time flow, head, voltage, current and temperature, the current pump efficiency is dynamically estimated. The difference between the rated operating conditions and the current operating conditions is combined to accurately calculate the power range within which the pump can be adjusted up or down in real time, thereby accurately evaluating the pump's online adjustment potential.

[0057] Step 2: Construct an ANN-based blower online adjustment capability model.

[0058] The model uses the blower's real-time speed, load torque, current, voltage, and ambient temperature as inputs and dynamically predicts the blower motor's real-time efficiency through an ANN model. By comparing the real-time prediction results with the rated load conditions, the model accurately quantifies the potential for power adjustments under the device's current state, enabling a precise assessment of the blower's regulation capabilities.

[0059] Step 3: Construct an evaluation model for the online adjustment capability of dosing equipment.

[0060] The dosing equipment regulation capacity assessment model uses the equipment's operating status as a basis for real-time identification of the dosing pump's on / off status. When the equipment is on, its full load is evaluated as the potential for downward adjustment, while when the equipment is off, the rated load is used as the potential for upward adjustment. This model clearly defines the relationship between the real-time on / off status and the equipment's rated power, enabling a dynamic and accurate assessment of its online power regulation potential.

[0061] Step 4: Construct an online adjustment capacity evaluation model for sludge dewatering equipment.

[0062] This model dynamically determines the equipment's current power adjustment potential based on the equipment's real-time operating status (operating or shut down). When the equipment is operating, the current power is used as the potential for downward adjustment; when the equipment is shut down, the rated power is used as the potential for upward adjustment. This allows for a clear and effective estimation of the equipment's real-time online adjustment capabilities.

[0063] Step 5: Construct and propose a lighting equipment online adjustment capability evaluation model.

[0064] The lighting equipment online adjustment capability model determines the power adjustment range of each lighting fixture based on its real-time on / off status. By monitoring lighting equipment operating data in real time, it determines whether the fixture has on / off or power adjustment capabilities. Based on this data, it summarizes the overall load adjustment range, providing a reliable basis for refined power control within the factory.

[0065] Step 6: Construct an online regulation capacity evaluation model for the water plant.

[0066] An online regulation capacity assessment model for integrated water pumps, blowers, dosing equipment, sludge dewatering equipment, and lighting equipment is used to evaluate the overall up / down regulation capacity of the water plant and respond to the online regulation needs of the power grid.

[0067] In this embodiment, for step 1, an artificial neural network is used to predict the pump efficiency in real time, and the collected information is shown in the following formula:

[0068] X=[x1,x2,x3,x4,x5] T

[0069] Where: x1 is the pump flow rate, x2 is the pump head, x3 is the input power, x4 is the pump speed, and x5 is the pump temperature.

[0070] The hidden layer of the ANN-based water pump online regulation capacity evaluation model is L layers, and the output h of the lth layer is (l) As shown in the following formula:

[0071] h (l) =f (l) (W( l) h (l-1) +b (l) ) l=1,2,…,L

[0072] Where: h (l) is the output of layer l, W (l) is the weight matrix of the lth layer, b (l) is the bias term of the lth layer, f (l) is a nonlinear activation function (such as ReLU, tanh or sigmoid function).

[0073] The output layer of the ANN-based water pump online regulation capability evaluation model is the efficiency prediction value. The mechanical efficiency prediction formula of the water pump is shown as follows:

[0074]

[0075] Where: η p is the predicted value of the mechanical efficiency of the water pump, W p is the output layer weight vector, b p is the output layer bias term, σ(·) is the Sigmoid function or other functions suitable for the output range [0,1], ensuring that the efficiency value is reasonable.

[0076] In this embodiment, the ANN-based water pump online regulation capability evaluation model is trained using historical operation data to minimize prediction errors.

[0077] Specifically, taking the mean square error (MSE) as the loss function, the water pump efficiency loss function is shown as follows:

[0078]

[0079] Where: Loss p is the pump efficiency loss function, The actual measurement efficiency of history, is the model prediction efficiency and N is the number of training samples.

[0080] The model optimizes parameters through the gradient descent method, and its calculation formula is shown as follows:

[0081]

[0082] Where: W (l) is the weight matrix of the lth layer, b (l) is the bias term of the lth layer, Loss is the loss function, and α is the learning rate of the model.

[0083] In real-time operation, the ANN-based water pump online regulation capacity evaluation model is used to predict the efficiency online, and the trained ANN model is used to calculate the real-time efficiency. The real-time efficiency calculation of the water pump is shown in the following formula.

[0084] η p (t) = ANNp (x1(t),x2(t),x3(t),x4(t),x5(t))

[0085] Where: η p For the real-time efficiency of the water pump, ANN p is the trained neural network model, x1 is the pump flow rate, x2 is the pump head, x3 is the input power, x4 is the pump speed, and x5 is the pump temperature.

[0086] In the water pump online regulation capacity evaluation model, the water pump load regulation capacity depends on the current operating flow and the allowable regulation range. Its upward regulation capacity and downward regulation capacity are shown in the following formulas:

[0087]

[0088] Where: The pump's upward adjustment capacity, is the pump's down-regulation capacity, Q(t) is the pump's current operating flow rate (m 3 / s), Q max is the maximum allowable flow rate (m 3 / s), Q min is the minimum stable flow rate (m 3 / s), ρ is the density of water kg / m 3 , g is the acceleration due to gravity m / s 2 ,η m is the motor efficiency, η p is the mechanical efficiency of the water pump, and H is the water pump head.

[0089] In this embodiment, for step 2, a blower online adjustment capability model is constructed. The model takes the blower real-time speed, load torque, current, voltage and ambient temperature as inputs, and dynamically predicts the real-time efficiency of the blower motor through an ANN model.

[0090] The real-time efficiency of the motor is calculated as follows:

[0091] η m (t) = ANN m (y1(t),y2(t),y3(t),y4(t),y5(t))

[0092] Where: η m Real-time efficiency of the motor, ANN m is the trained neural network model, y1 is the real-time speed of the blower, y2 is the load torque of the blower, y3 is the current of the blower, y4 is the voltage of the blower, and y5 is the ambient temperature.

[0093] In the ANN-based blower online adjustment capacity evaluation model, the equipment adjustment capacity depends on the current load and the allowable variation range. Its upward and downward adjustment capabilities can be calculated by the following formula:

[0094]

[0095] Where: For the blower's upward adjustment capability, is the blower's down-regulation capability, T(t) is the current load torque (N·m), T max is the maximum allowable torque of the load (N·m), T min is the minimum allowable torque of the load (N·m), η m is the motor efficiency and n is the motor speed.

[0096] In this embodiment, for step 3, an online adjustment capability model of the dosing equipment is constructed. The dosing equipment adjustment capability evaluation model is based on the equipment operation status and identifies the start and stop status of the dosing pump in real time.

[0097] Specifically, when the equipment is turned on, its entire load is evaluated as the downward adjustment space, and when the equipment is turned off, the rated load is used as the adjustable power. The relationship between the real-time switching status and the rated power of the equipment is clarified to dynamically and accurately evaluate its online power regulation potential.

[0098] Dosing equipment typically operates intermittently, with low power consumption, but it does have a certain degree of on / off regulation capability. Assuming the dosing equipment operates in an on / off mode: when the equipment is on, it has the ability to reduce the full load; when the equipment is off, it has the ability to increase the full load.

[0099] The upward and downward adjustment capabilities of the dosing equipment are shown in the following formula.

[0100]

[0101] Where: To increase the capacity of the dosing equipment, For the down-regulation capacity of the dosing equipment, X d (t) is the operating status of the device (0 is off, 1 is running), P d,max It is the maximum rated power of the dosing equipment (W).

[0102] In this embodiment, for step 4, an online adjustment capability model of the sludge dewatering equipment is constructed. The model dynamically determines the current power adjustment space of the equipment according to the real-time operating status (operating or shut down) of the sludge dewatering equipment.

[0103] Specifically, when the equipment is running, the current power is used as the potential for downward adjustment; when the equipment is stopped, the rated power is used as the potential for upward adjustment. This allows for a clear and effective estimation of the real-time online adjustment capability of the sludge dewatering equipment. Sludge equipment is generally adjusted discretely: there is downward adjustment capability when the equipment is running and upward adjustment capability when the equipment is stopped.

[0104] The upward and downward adjustment capabilities of the sludge equipment can be calculated by the following formula:

[0105]

[0106] in: To increase the capacity of sludge equipment, For the down-regulation capacity of sludge equipment, X s (t) is the equipment operation status (0 is stopped, 1 is running), P s,max The maximum power rating of the device (W).

[0107] In this embodiment, for step 5, an online adjustment capability model of the lighting device is constructed. The online adjustment capability model of the lighting device determines the power adjustment range of each lighting device based on the real-time on / off status of the lighting device.

[0108] Specifically, by using real-time monitoring of lighting equipment operating data, it is determined whether the lighting equipment has a switching or power adjustment function, and based on this, the upward and downward adjustment space of the overall load is summarized, providing a reliable basis for refined power control within the factory.

[0109] The up / down adjustment capability of a lighting device is as follows:

[0110]

[0111] Where: For lighting equipment up / down ability, X li (t) is whether the i-th lamp is adjustable (0 is not adjustable, 1 is adjustable), P unit,i is the rated power of the i-th lamp (W), n l is the number of lighting devices.

[0112] In this embodiment, for step 6, an online regulation capability model of the water plant is constructed.

[0113] In this embodiment, the online adjustment capability evaluation implementation process is as follows:

[0114] First, the operating data of various equipment (flow rate, power, speed, etc.) is collected in real time, and the online adjustment capacity of each equipment is calculated in real time based on the rated parameters of each equipment;

[0115] Then, the online regulation capability of each device is summarized to obtain the overall online regulation capability of the water plant;

[0116] Finally, the real-time change trend of the online regulation capability is evaluated to provide a basis for load regulation decisions.

[0117] Specifically, the online regulation capacity model of the water plant can be calculated by the following formula:

[0118]

[0119] Where: To improve overall capacity, For the overall downward adjustment capacity, The pump's upward adjustment capacity, For the blower's upward adjustment capability, To increase the capacity of the dosing equipment, To increase the capacity of sludge dewatering equipment, For the upward adjustment capability of lighting equipment, The pump's down-regulation capability, The blower's down-regulation capability, For the down-regulation capacity of dosing equipment, To reduce the capacity of sludge dewatering equipment, Down-regulation capability for lighting equipment

[0120] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A method for evaluating the online regulation capacity of a waterworks based on ANN, characterized in that: The steps include: Step 1: Construct an ANN-based water pump online adjustment capability evaluation model: Using an artificial neural network, input real-time flow, head, voltage, current, and temperature parameters to dynamically predict the current water pump efficiency. Combining the difference between the rated operating conditions and the current operating conditions, the power range that the water pump can adjust up or down in real time is calculated to evaluate the water pump's online adjustment potential. Step 2: Build an ANN-based online blower adjustment capability model: Using the blower's real-time speed, load torque, current, voltage, and ambient temperature as input, the artificial neural network dynamically predicts the current blower motor efficiency. By comparing the difference between the predicted results and the rated load conditions, the power adjustment space of the blower in the current state is quantified, and the blower's adjustment capability is evaluated. Step 3: Build an online adjustment capability assessment model for dosing equipment: Identify the on / off status of the dosing pump in real time, evaluate the full load as the downward adjustment space when it is on, and use the rated load as the upward adjustment space when it is off. Determine the relationship between the real-time on / off status and the rated power of the dosing equipment, and dynamically evaluate the online power adjustment potential of the dosing equipment. Step 4. Construct an online adjustment capacity evaluation model for the sludge dewatering equipment: Based on the real-time operating status of the sludge dewatering equipment, dynamically determine the current power adjustment space of the sludge dewatering equipment. When the equipment is in operation, the current power is used as the downward adjustment potential. When the equipment is in shutdown, the rated power is used as the upward adjustment potential. The real-time online adjustment capacity of the sludge dewatering equipment is estimated. Step 5: Build a lighting equipment online adjustment capability assessment model: Determine the power adjustment range based on the real-time on / off status of each lighting fixture. Leverage real-time monitoring of lighting equipment operating data to determine whether the fixtures have on / off or power adjustment capabilities. This model also summarizes the overall load adjustment potential for refined power control within the factory. Step 6. Construct an online regulation capacity assessment model for the waterworks: Build an online regulation capacity assessment model for integrated water pumps, blowers, dosing equipment, sludge dewatering equipment, and lighting equipment to assess the overall regulation up / down capability of the waterworks and respond to the online regulation needs of the power grid.

2. The method for evaluating the online regulation capability of a waterworks according to claim 1, wherein: In step 1, the artificial neural network is used to predict the pump efficiency in real time. The formula of the collected information is as follows: X=[x1,x2,x3,x4,x5] T Where x1 is the pump flow rate, x2 is the pump head, x3 is the input power, x4 is the pump speed, x5 is the pump temperature, and the superscript T indicates transposition; The water pump online regulation capacity evaluation model based on ANN has L hidden layers and the output h of the lth layer is (l) The formula is as follows: h (l) =f (l) (W (l) h (l-1) +b (l) ) l=1,2,…,L Where h (l) is the output of layer l, W (l) is the weight matrix of the lth layer, b (l) is the bias term of the lth layer, f (l) is a nonlinear activation function; The ANN-based water pump online regulation capability evaluation model has an output layer that is the efficiency prediction value. The mechanical efficiency prediction formula of the water pump is as follows: Where η p is the predicted value of the mechanical efficiency of the water pump, W p is the output layer weight vector, b p is the output layer bias term, and σ(·) is a function of the output range [0,1].

3. The method for evaluating the online regulation capability of a waterworks according to claim 2, wherein: The ANN-based water pump online regulation capacity evaluation model is trained using historical operation data to minimize prediction errors; Taking mean square error as loss function, the water pump efficiency loss function is expressed as follows: Where, Loss p is the pump efficiency loss function, The actual measurement efficiency of history, is the model prediction efficiency, N is the number of training samples; The parameters are optimized by gradient descent method, and the calculation formula is as follows: Where, is W (l) is the weight matrix of the lth layer, b (l) is the bias term of the lth layer, Loss is the loss function, and α is the learning rate of the model.

4. The method for evaluating the online regulation capability of a waterworks according to claim 3, wherein: The real-time efficiency of the water pump is calculated using the trained water pump online regulation capability evaluation model. The formula is as follows: η p (t)=ANN p (x1(t),x2(t),x3(t),x4(t),x5(t)) Where: η p For the real-time efficiency of the water pump, ANN p is the trained neural network model, x1 is the pump flow rate, x2 is the pump head, x3 is the input power, x4 is the pump speed, and x5 is the pump temperature; The adjustment capability of the pump's real-time load is determined based on the current operating flow and the allowable adjustment range. The formula is as follows: Where: The pump's upward adjustment capacity, is the pump's down-regulation capacity, Q(t) is the pump's current operating flow rate, and Q max is the maximum allowable flow rate, Q min is the minimum stable flow rate, ρ is the density of water, g is the acceleration due to gravity, η m is the motor efficiency, η p is the mechanical efficiency of the water pump, and H is the water pump head.

5. The method for evaluating the online regulation capability of a waterworks according to claim 1, wherein: In the blower online adjustment capability model described in step 2, the real-time efficiency of the motor is calculated using the following formula: η m (t)=ANN m (y1(t),y2(t),y3(t),y4(t),y5(t)) Where: η m Real-time efficiency of the motor, ANN m is the trained neural network model, y1 is the real-time speed of the blower, y2 is the load torque of the blower, y3 is the current of the blower, y4 is the voltage of the blower, and y5 is the ambient temperature; Determine the online adjustment capability of the blower equipment based on the current load and the allowable variation range. The calculation is as follows: Where: For the blower's upward adjustment capability, is the blower's down-regulation capability, T(t) is the current load torque, T max is the maximum torque allowed by the load, T min is the minimum torque allowed by the load, η m is the motor efficiency and n is the motor speed.

6. The method for evaluating the online regulation capability of a waterworks according to claim 1, wherein: In the online adjustment capability model of the dosing equipment described in step 3, the dosing equipment operates intermittently and has on / off adjustment capability; When the dosing equipment is turned on, it has the ability to adjust the entire load downward; when the dosing equipment is turned off, it has the ability to adjust the entire load upward. The upward and downward adjustment capabilities of the dosing equipment are calculated as follows: Where: To increase the capacity of the dosing equipment, For the down-regulation capacity of the dosing equipment, X d (t) is the operating status of the dosing equipment, P d,max The maximum rated power of the dosing equipment.

7. The method for evaluating the online regulation capability of a waterworks according to claim 1, wherein: In the online regulation capacity model of the sludge equipment described in step 4, the sludge equipment is discretely regulated, with downward regulation capacity when running and upward regulation capacity when stopped. The upward and downward regulation capacities of the sludge equipment are calculated as follows: in: To increase the capacity of sludge equipment, For the down-regulation capacity of sludge equipment, X s (t) is the operating status of the sludge equipment, P s,max The maximum power rating of the device.

8. The method for evaluating the online regulation capability of a waterworks according to claim 1, wherein: In the lighting equipment online adjustment capability model described in step 5, the lighting equipment's up / down adjustment capability is calculated as follows: Where: ΔP l up / down (t) is the up / down adjustment capability of the lighting equipment, X li (t) is whether the i-th lamp is adjustable, P unit,i is the rated power of the i-th lamp, n l is the number of lighting devices.

9. The method for evaluating the online regulation capability of a waterworks according to claim 1, wherein: The online regulation capability model of the water plant described in step 6 and the specific steps for online regulation capability evaluation are as follows: Step 6.1: Real-time collection of operating data of various equipment in the water plant, including flow rate, power, and speed; and real-time calculation of the online adjustment capacity of various equipment based on the rated parameters of the equipment; Step 6.2: Summarize the online regulation capabilities of various equipment to obtain the overall online regulation capability of the water plant; Step 6.3: Evaluate the real-time trend of online regulation capability to provide a basis for load regulation decisions.

10. The method for evaluating the online regulation capability of a waterworks according to claim 9, characterized in that: The online regulation capacity model of the water plant summarizes the online regulation capacity of various equipment and is calculated as follows: Where: To improve overall capacity, For the overall downward adjustment ability, The pump's upward adjustment capacity, For the blower's upward adjustment capability, To increase the capacity of the dosing equipment, The upward adjustment capacity of the sludge dewatering equipment, ΔP l up (t) is the upward adjustment capability of the lighting equipment, The pump's down-regulation capability, The blower's down-regulation capability, For the down-regulation capacity of dosing equipment, The downward adjustment capacity of the sludge dewatering equipment, ΔP l down (t) is the downward adjustment capability of the lighting equipment.

Images