Operation support equipment for hydroelectric power plants, and method for supporting the operation of hydroelectric power plants.

Abstract

This enables a reduction in computational load while suppressing a decline in the accuracy of hydroelectric power plant operation plans. [Solution] An operational support device for a hydroelectric power plant according to an embodiment of the present invention comprises an evaluation function generation unit and an optimization calculation processing unit. The evaluation function generation unit generates an evaluation function based on the power output of a generator that changes over time according to the time-series amount of water used for power generation drawn from the reservoir of the dam. The optimization calculation processing unit constrains the relationship between the time-series amount of water used for power generation and the time-series power output relative to the time-series amount of water used for power generation to a predetermined operating point that changes according to the head based on the reservoir of the dam, and generates a planned value for the time-series amount of water used for power generation so as to maximize the evaluation function.

Description

[Technical Field]

[0001] Embodiments of the present invention relate to an operational support device for a hydroelectric power plant and an operational support method for a hydroelectric power plant. [Background technology]

[0002] The amount of water flowing into a dam system generally varies depending on factors such as rainfall upstream, snowmelt, and spring water. In water system management, it is necessary to plan and operate the efficient water usage of connected power plants in accordance with the amount of water flowing into the dam system in order to obtain power generation in proportion to the inflow.

[0003] In such river systems, dam-type power plants have a change in the head of the power generation equipment depending on the dam water level. High water level operation can increase power generation, but when the dam water level reaches its upper limit, it may lead to a decrease in power generation due to the operation of releasing water from the dam that is not used for power generation.

[0004] Regarding the operational planning of hydroelectric power plants that takes into account the difference in head due to changes in dam water levels, there are known methods for creating a power generation operational plan for hydroelectric power plants by repeatedly updating simulations of changes in water storage volume using a power output model that takes into account changes in head (see, for example, Patent Document 1), and operational plan generation devices that create an operational plan for hydroelectric power plants by dividing the range of possible water storage volume and discharge volume of the power plant and repeatedly generating an optimization problem and deriving the optimal solution (see, for example, Patent Document 2). [Prior art documents] [Patent Documents]

[0005] [Patent Document 1] Patent No. 5425985 [Patent Document 2] Patent No. 7362793 [Overview of the project] [Problems that the invention aims to solve]

[0006] However, in order to ensure the accuracy of hydroelectric power plant operation plans through iterative calculations, there is a trade-off between computational load and planning accuracy, requiring high-load computational processing. Furthermore, in operation plans for dam water systems where multiple power plants are connected, the computational load becomes even higher, sometimes making practical operation difficult.

[0007] Embodiments of the present invention have been made in consideration of these circumstances, and provide a hydroelectric power plant operation support device and a hydroelectric power plant operation support method that can reduce the computational load while suppressing a decrease in the accuracy of the operation plan of the hydroelectric power plant. [Means for solving the problem]

[0008] An operational support device for a hydroelectric power plant according to an embodiment of the present invention comprises an evaluation function generation unit and an optimization calculation processing unit. The evaluation function generation unit generates an evaluation function based on the power output of a generator that changes over time according to the time-series amount of water used for power generation drawn from the dam's reservoir. The optimization calculation processing unit constrains the relationship between the time-series amount of water used for power generation and the time-series power output relative to the time-series amount of water used for power generation to a predetermined operating point that changes according to the head based on the dam's reservoir, and generates a planned value for the time-series amount of water used for power generation to maximize the evaluation function. [Effects of the Invention]

[0009] According to the present invention, it is possible to reduce the computational load while suppressing the deterioration of the accuracy of the operation plan for hydroelectric power plants. [Brief explanation of the drawing]

[0010] [Figure 1] A block diagram showing an example configuration of the operation support system for a hydroelectric power plant according to this embodiment. [Figure 2] A diagram showing an example of a model of a river system that is subject to a river system management plan. [Figure 3] A block diagram showing an example configuration of the processing unit according to this embodiment. [Figure 4] A diagram showing pipeline loss characteristic data. [Figure 5]A diagram showing turbine efficiency as turbine characteristic data. [Figure 6] A diagram showing generator efficiency as generator characteristic data. [Figure 7] A flowchart illustrating an example of the process for generating hydroelectric power generation data. [Figure 8] This figure shows water-based power generation data illustrating the relationship between power output and water usage for power generation. [Figure 9] This diagram shows water-based power generation data illustrating the relationship between power output and water usage for power generation in the case of two units. [Figure 10] This figure shows an example of a computational model generated by the luck evaluation model generation unit. [Figure 11] A diagram showing the relationship between various decision variables at different time steps. [Figure 12] A table illustrating the selection of operating points. [Figure 13] A flowchart illustrating an example of processing for an operational support system at a hydroelectric power plant. [Modes for carrying out the invention]

[0011] Hereinafter, a power conversion device and control method according to embodiments of the present invention will be described in detail with reference to the drawings. Note that the embodiments shown below are examples of embodiments of the present invention, and the present invention is not limited to these embodiments. Furthermore, in the drawings referenced in these embodiments, the same or similar reference numerals are used for identical parts or parts having similar functions, and repeated descriptions may be omitted. Also, some components may be omitted from the drawings.

[0012] (One embodiment)

[0013] Figure 1 is a block diagram showing an example configuration of the hydroelectric power plant operation support system 1 according to this embodiment. The operation support system 1 according to this embodiment is a system capable of supporting the operation of, for example, generators P1 to Pn.

[0014] The operation support system 1 according to this embodiment comprises an operation support device 10 for a hydroelectric power plant, a display device 20, an input device 30, and a prediction device 40. The operation support device 10 for the hydroelectric power plant has a calculation processing unit 12 and an operation characteristic data storage unit 14. Figure 1 also schematically illustrates each dam Di (1 ≤ i ≤ n). The subscript "i" is a subscript attached to the variable of each dam or each power plant. n is a natural number and can be set to any number.

[0015] Furthermore, the devices related to these operational support systems 1 may not be located at a single location but rather distributed across multiple locations, and the system may be configured to communicate with each device located at a remote location via a communication system. Also, in the following explanation, when describing events common to each dam or power plant, the symbol "i" may be omitted.

[0016] Each dam Di (1≦i≦n) is equipped with a generator Pi, a water level meter WLi, an inflow meter Iqi, an outflow meter Oqi, and the like. The hydroelectric power plant's operation support device 10, prediction device 40, generator Pi, water level meter WLi, inflow meter Iqi, and outflow meter Oqi are configured to exchange signals containing various information with each other via a network nw. As a result, information such as the rotational speed of each generator Pi (1≦i≦n), the amount of water supplied, the amount of power generated, the total head, and the amount of water flowing into each dam Di (1≦i≦n) is supplied to the hydroelectric power plant's operation support device 10 and prediction device 40. Hereinafter, the hydroelectric power plant's operation support device 10 may be simply referred to as the operation support device 10.

[0017] A generator Pi is, for example, a hydroelectric generator, and may consist of multiple generators. That is, a power plant Ppi has at least one generator Pi.

[0018] The operation support device 10 is composed of, for example, a CPU (Central Processing Unit), an MPU (Micro Processor Unit), RAM, and ROM, and each processing unit is configured by executing programs stored in RAM and ROM. This operation support device 10 is a device that can provide information on methods for planning water system operations.

[0019] The calculation processing unit 12 of the operation support device 10 calculates, for example, the operation plan for the amount of water used for power generation at each power plant Ppi (1≦i≦n) and the discharge rate at each dam Di (1≦i≦n) in a time series, based on the operating characteristic data of each power plant Ppi (1≦i≦n), the inflow prediction data of each dam Di (1≦i≦n), and the constraint conditions of each dam Di (1≦i≦n). Furthermore, the operation support device 10 of the hydroelectric power plant can visualize the operation plan in a chart and display it on the display device 20. Details of the calculation processing unit 12 will be described later with reference to Figure 3.

[0020] The operating characteristics data storage unit 14 is composed of, for example, an HDD (hard disk drive) or an SSD (solid state drive). As described above, this operating characteristics data storage unit 14 stores information such as the rotational speed of each generator Pi (1≦m≦n), the amount of water supplied, the amount of power generated, the water level, and the amount of water flowing into each dam Di (1≦i≦n) in a time series. It also stores the operating characteristics data of each power plant Ppi (1≦i≦n).

[0021] The operational characteristics data stores pipeline loss characteristics data, generator characteristics data, and turbine characteristics data for each power plant Ppi (1 ≤ i ≤ n). Furthermore, it stores water flow power generation data showing the relationship between water usage and power output, calculated using this data. In addition, it stores high-efficiency route data and high-output route data generated using this water flow power generation data. Details of this data will be described later.

[0022] The display device 20 is, for example, a monitor. This display device 20 displays image information generated by the operation support device 10.

[0023] The input device 30 includes, for example, a mouse and a keyboard. The input device 30 inputs operation signals according to the operator's instructions to the hydroelectric power plant's operation support device 10.

[0024] The prediction device 40 uses weather forecasts, rainfall information, and snow depth information supplied via the network nw from the Japan Meteorological Agency's server 50 to predict the time-series inflow of water into the water system. More specifically, the prediction device 40 stores in its memory the time-series data of past weather forecasts, rainfall information, and snow depth information in association with historical inflow data to each dam Di (1≦i≦n). For example, it stores several years' worth or more of this historical data.

[0025] The prediction device 40 uses this historical data to train a prediction model that takes weather forecasts, rainfall information, and snow depth information as input and outputs the inflow qin(1≦i≦n) to each dam Di(1≦i≦n). For this type of training, general neural network learning models, multivariate analysis models such as linear regression, etc., can be used.

[0026] The forecasting device 40 is supplied with time-series forecast data of weather forecasts, rainfall information, and snow depth information for three days, for example, in 30-minute intervals. This allows the forecasting device 40 to use a forecasting model to output three days' worth of forecast values ​​of the water inflow rate qini (1≦i≦n) into each dam Di (1≦i≦n) to the operation support device 10 in time-series intervals of 30 minutes.

[0027] Alternatively, the prediction device 40 uses weather forecasts, rainfall information, and snowfall information supplied via the network nw to refer to past weather forecasts, past rainfall information, etc. stored in the memory unit, as well as to refer to actual data on past inflow amounts for each dam stored in the memory unit, and outputs predicted values ​​of the water inflow amount qini (1≦i≦n) to each dam in a time series.

[0028] Here, instead of using actual data on the inflow of each dam, it is also possible to input actual data on the discharge volume, water intake for power generation, and water level changes of each dam, and use the calculated inflow for each dam. Alternatively, the prediction of the inflow for each dam can also be done by predicting the flow rate of the rivers flowing into each dam.

[0029] (Water system model) The water system model according to this embodiment will be explained using Figure 2. Figure 2 is a diagram showing an example of a water system model that is the target of a water system management plan.

[0030] The model shown in Figure 2 is an example of a model for a dam system in which multiple dams D1, D2, and D3 are connected. For simplicity, we will explain using a three-stage dam example here, but it is not limited to this, and an i-stage dam would also be acceptable. As mentioned above, i is in the range of 1 ≤ i ≤ n, and n is any natural number.

[0031] Dams D1, D2, and D3 are equipped with generators P1, P2, and P3, respectively, which generate electricity using water taken from each dam. Generators P2 and P3 are examples of systems with two generators. The amount of water used for power generation q1-q3 after use by the generators at each dam is sent to the next dam along with the amount of water qd1-qd3 released from the dam gates of each dam D1-D3.

[0032] Let qin1, qin2, and qin3 be the inflow rates of water into dams D1, D2, and D3, respectively. Let h1, h2, and h3 be the total heads of dams D1, D2, and D3, respectively. Let v1, v2, and v3 be the water storage capacities of dams D1, D2, and D3, respectively. Let p1, p2, and p3 be the power generation amounts of generators P1, P2, and P3, respectively. Note that the power generation amounts of generators P1, P2, and P3 are sometimes referred to as power output or turbine power output.

[0033] The arithmetic processing unit 12 according to this embodiment will be described using Figure 3. Figure 3 is a block diagram showing an example of the configuration of the arithmetic processing unit 12 according to this embodiment, and it is possible to generate an operation plan using a water system model such as Figure 2 and Figure 10, which will be described later.

[0034] The arithmetic processing unit 12 performs various arithmetic operations. This arithmetic processing unit 12 comprises an acquisition unit 100, a characteristic data generation unit 102, a control unit 104, and an operation plan calculation unit 106.

[0035] The acquisition unit 100 is the input interface between the network nw and the input device 30. The acquisition unit 100 acquires information such as the rotational speed, water supply amount, power generation amount, water level, and inflow amount to each dam Di (1≦i≦n) of each generator Pi (1≦i≦n) via the network nw, associating them with time series, and stores them in the operating characteristics data storage unit 14. In addition, it acquires operating characteristics data of each power plant Ppi (1≦i≦n) via the network nw and stores it in the operating characteristics data storage unit 14.

[0036] [Processing by the characteristic data generation unit] The characteristic data generation unit 102 generates pipeline loss characteristic data (see Figure 4 below), turbine characteristic data (see Figure 5 below), and generator characteristic data (see Figure 7 below) if these cannot be obtained via the network nw. Furthermore, the characteristic data generation unit 102 uses the pipeline loss characteristic data, turbine characteristic data, and generator characteristic data to generate water flow power generation data (see Figures 8 and 9 below), which shows the relationship between the amount of water used for power generation and the power output. In addition, the characteristic data generation unit 102 generates high-efficiency route data (see Figures 8 and 9 below) and high-output route data (see Figures 8 and 9 below) using the water flow power generation data. Details of the characteristic data generation unit 102 will be described later using Figures 4 to 9.

[0037] The control unit 104 controls the entire operation support system 1. The control unit 104 also displays charts, numerical values, etc., generated by the operation plan calculation unit 106 (described later) on the display device 20.

[0038] The operation planning calculation unit 106 calculates the amount of inflow q to each dam Di (1≦i≦n) supplied via the network nw. ini Using the predicted data (t)(1≦i≦n), the amount of water used for power generation q for each power plant Ppi(1≦i≦n) is calculated. i (t)(1≦i≦n), discharge amount q di (t)(1≦i≦n), power generation p i (t)(1≦i≦n) and other information, at least the amount of water used for power generation q i Calculate an operational plan that includes (t)(1≦i≦n). Here, t represents time.

[0039] Inflow rate q ini (t)(1≦i≦n) The prediction data may be data generated by the prediction device 40, or may be data supplied from another device via the network nw.

[0040] This operation plan calculation unit 106 includes an evaluation model generation unit 108, an evaluation function generation unit 110, a constraint condition generation unit 112, an optimization calculation processing unit 114, and a plan generation unit 116. The details of the operation plan calculation unit 106 will also be described later.

[0041] Here, the details of the characteristic data generation unit 102 will be described. FIG. 4 is a diagram showing pipeline loss characteristic data. The horizontal axis is the power generation water consumption q [m 3 / s], and the vertical axis is the pipeline pressure loss (pipeline loss head) l [m]. The pipeline pressure loss l [m] indicates, for example, the pressure loss due to the slope of the water conduit or the friction of the penstock. Line L10 indicates the pipeline pressure loss l [m] in the case of single-unit operation, and line L20 indicates the pipeline pressure loss l [m] in the case of two-unit operation. The relationship between the power generation water consumption q and the turbine flow rate q t is q = N·q t where N indicates the number of operating units. In this embodiment, units are used exemplarily, but are not limited thereto. For example, arbitrary units can be used for the units according to this embodiment.

[0042] The pipeline pressure loss l [m] can be generated, for example, using the inflow loss coefficient of the pipeline, the flow velocity [m / s] before inflow into the pipeline, and the flow velocity [m / s] after inflow into the pipeline. Thus, the characteristic data generation unit 102 can generate pipeline loss data l [m] for the case of one generator and the case of two generators.

[0043] The characteristic data generation unit 102 generates an effective head (h - l) [m] by subtracting the pipeline pressure loss l [m] from the total head h [m]. The total head h [m] indicates, for example, the elevation difference between the intake level (the water surface elevation at the intake point) and the discharge level (the water surface elevation where water is discharged from the power plant).

[0044] Figure 5 shows turbine efficiency η as turbine characteristic data. t This figure shows the percentages [%]. The horizontal axis represents the rotational speed per unit head n / Sqrt(hl), and the vertical axis represents the flow rate per unit head q. t The value is / Sqrt(hl). The flow rate per unit head is the turbine flow rate q. t [m 3 This is the value obtained by dividing [ / s] by the square root of the effective head (hl) [m]. In other words, the turbine efficiency η t [%] represents the ratio of turbine output to theoretical power. Also, the amount of water used for power generation q[m³] 3 The value of [ / s] is given by equation (1), where N is the number of generators used for power generation.

number

[0045] Figure 6 shows the generator efficiency η as generator characteristic data. g This figure shows the percentages [%]. The horizontal axis represents the load factor p. t / p tR The vertical axis represents the generator efficiency η. g [%]. Load factor p t / p tR p is the rated value of the turbine output. tR and turbine power output p t This is the ratio to the turbine power output p. t This is expressed by equation (2), where g is the acceleration due to gravity and ρ is the density of water.

number

[0046] Figure 7 is a flowchart showing an example of the water-power generation data generation process of the characteristic data generation unit 102, illustrating the relationship between the amount of water used for power generation and the power output. The power output p [MW] of each generator Pi (1 ≤ i ≤ n) is shown by equation (3). N is the number of generators used for power generation.

number

[0047] As shown in Figure 7, the characteristic data generation unit 102 uses the pipeline loss characteristic data (see Figure 4), turbine characteristic data (see Figure 5), and generator characteristic data (see Figure 6) of each generator Pi (1≦i≦n) to generate the power output p [MW] and the amount of water used for power generation q [m³] for each generator Pi (1≦i≦n). 3 Characteristic data (Figures 8 and 9 described later) showing the relationship with / s] is generated. That is, the characteristic data generation unit 102 uses the characteristic data for each dam Di (1 ≤ i ≤ n) to generate power output pi (1 ≤ i ≤ n) [MW] and power generation water usage qi (1 ≤ i ≤ n) [m 3 This generates characteristic data (Figures 8 and 9 described later) that shows the relationship with [ / s].

[0048] First, the characteristic data generation unit 102 obtains the total head h [m] within the possible range (step S100). Subsequently, it obtains the total head h [m] and the amount of water used for power generation q [m 3 The input values ​​are shifted (step S102) and the calculation of the power output p[MW] (see equation (3)) is repeated (step S103).

[0049] Figure 8 shows water-based power generation data illustrating the relationship between power output and water usage. Figure 8 is an example of characteristic data for a single generator (N=1). The horizontal axis represents the water usage q[m 3 The vertical axis is the power output p[MW] (see equation (1)), and the vertical axis is the power output p[MW] (see equation (3)). As shown in Figure 8, the characteristic data generation unit 102 generates the minimum value h of the total head h L [m] to the maximum value h h Within the range [m], calculate the power output p for the amount of water used for power generation q.

[0050] Furthermore, the characteristic data generation unit 102 generates characteristic data showing the relationship between power output p and water usage q, and then generates high-efficiency route Lq1 and maximum output route Lq2 for each total head h. High-efficiency route Lq1 is the highest-efficiency operating point where the ratio of power output p to water usage q is maximized, for example (q 1L , p 1L ) and (q 1H , p HThis is the line connecting (q). In the case of maximizing the power-to-water ratio, which is the ratio of power output (amount of electricity generated) p [MW] to the amount of water used for power generation q, the highest efficiency operating point is used as a constraint. The maximum output path Lq2 is the power plant's maximum output operating point where the power output p is at its maximum value, for example (q 2L , p 2L ) and (q 2H , p 2H This is the line connecting ).

[0051] Figure 9 shows water-based power generation data illustrating the relationship between power output and water usage for power generation when there are two generators. The horizontal axis represents the water usage q[m³] for power generation. 3 The equation is [ / s] (see equation (1)), and the vertical axis is the power output p [MW] (see equation (3)). The example of power generation with one unit (N=1) up to the maximum output path Lq2 is shown, and the example of power generation with two units (N=2) in the range where more water usage q is used for power generation than the maximum output path Lq2 is shown.

[0052] The high-efficiency path Lq3 is the point of highest efficiency when generating power with two units (N=2), where the ratio of power output p to the amount of water used for power generation q is maximized, for example (q 3L , p 3L ) and (q 3H , p 3H This is the line connecting (q). The maximum output path Lq4 is the power plant's maximum output operating point where the power output p is at its maximum value when generating power with two units (N=2), for example (q 4L , p 4L ) and (q 4H , p 4H This is a line connecting the generators. The characteristic data generation unit 102 stores characteristic data, including the high-efficiency paths Lq1i and Lq3i (1≦i≦n), and the maximum output paths Lq2i and Lq4i (1≦i≦n) for each generator Pi (1≦i≦n), in the operating characteristic data storage unit 14 (see Figure 1).

[0053] [Processing by the Operational Planning Calculation Unit] Referring again to Figure 3, we will explain the detailed processing of the operation plan calculation unit 106.

[0054] Figure 10 shows an example of a calculation model generated by the evaluation model generation unit 108 of the operation plan calculation unit 106. For simplicity of explanation, a two-tiered dam is shown here. In Figure 10, as mentioned above, i represents the dam number, in the range 1 ≤ i ≤ n. n is a natural number. In other words, Figure 10 shows a two-tiered dam as an example for 1 ≤ i ≤ 2 for simplicity of explanation, but is not limited to this. Note that the inflow q ini (t) is, for example, the predicted value from the prediction device 40 (see Figure 1).

[0055] Figure 11 shows the relationship between various decision variables and the dam water volume v at each time step Δt. i (t), total drop h i (t), amount of water used for power generation q i (t), power output p i (t) has the relationship shown in equations (1) to (5) above. As an example of the sampling Δt of the operation plan, let's assume it is 30 minutes. Therefore, when targeting a 2-day operation plan, nt = 96, i.e., t = 1, 2, ..., 96. Note that the time step Δt in this embodiment is, for example, 30 minutes, but is not limited to this.

[0056] The evaluation model generation unit 108 generates the inflow q into the dam Di. ini (t), dam reservoir v i (t), total drop h i (t), amount of water used for power generation q i (t), discharge amount q di (t), power output p i The relationship between (t) and is modeled. The evaluation model generation unit 108 generates the dam water volume v i (t) is modeled as a dam reservoir model using equation (4). For example, each unit example is the inflow q ini (t)[m 3 / s], dam water storage v i (t)[m 3 ], total drop h i (t)[m], amount of water used for power generation q i (t)[m 3 / s], discharge amount q di (t)[m 3 / s], power output pi (t)[MW]

number

[0057] Also, the dam's water storage volume v i (t) and total drop h i The dam characteristics that show the relationship with (t) are coefficient a i、 b i It is modeled by equation (5) using the following. The characteristic data generation unit 102 generates, for example, coefficient a i、 b i This is calculated in advance using linear regression analysis and stored in the driving characteristics data storage unit 14 (see Figure 1).

number

[0058] The evaluation function generation unit 110 generates the evaluation function H shown in equation (6). Power unit price c i The price (yen / kWh) can be obtained from the power company or electricity retailer that purchases electricity via the network. When selling electricity to the wholesale electricity market, the electricity unit price is calculated by predicting the electricity trading price. i (yen / kWh) may also be set. In other words, the evaluation function H shown in equation (6) is the evaluation function when the objective is to maximize the amount of electricity sold.

[0059] Also, electricity unit price c i If is set to a constant value, for example 1, then it becomes an evaluation function H that shows the total amount of electricity generated during the operational planning period of the target river system. That is, the evaluation function H shown in equation (6) is equal to the electricity unit price c i By setting this value to a constant, it becomes an evaluation function for maximizing the amount of electricity generated. Alternatively, the evaluation function H can be minimized by adding a negative sign.

number

[0060] The constraint generation unit 112 generates constraints for maximizing the evaluation function H, for example. Referring again to Figures 8 and 9, in this embodiment, the amount of water used for power generation q i (t) and power output p i The relationship with (t) is set to high-efficiency path Lq1, maximum output path Lq2, high-efficiency path Lq3, and maximum output path Lq4. This makes it possible to quickly calculate the evaluation function shown in equation (6) according to the purpose. For example, if the goal is maximum efficiency, high-efficiency paths Lq1 and Lq3 are used as constraints. On the other hand, if the goal is maximum output, maximum output paths Lq2 and Lq4 are used as constraints. Depending on the circumstances of the hydroelectric power plant in question, additional operating points may be added and introduced into the constraints. The following explanation will use an example where high-efficiency paths Lq1 and Lq3 and maximum output path Lq4 are used as constraints. This is the case when the high-efficiency operating point for each number of operating units in the power plant and the operating point for the power plant's maximum output are operational.

[0061] The constraint generation unit 112 generates the following constraint equations for each dam Di (1 ≤ i ≤ n). Equations (7) and (8) are examples of constraint equations when the high-efficiency paths Lq1 and Lq3 and the maximum output path Lq4 are selectively used. In other words, it is also possible to select the high-efficiency paths Lq1 and Lq3 and the maximum output path Lq4 for each dam Di (1 ≤ i ≤ n).

number

number

[0062]

number

[0063] Figure 12 is a table illustrating the selection of the operating point. As shown in Figure 12, z 0i (t), z 1i (t), z 2i (t), z 3i (t) represents 0 or 1. 0i When (t) is 1, power generation stops. 1iWhen (t) is 1, the high-efficiency path Lq1 is used as a constraint condition. That is, z 1i When (t) is 1, when the number of operating units is 1, the highest-efficiency operating point is used as a constraint.

[0064] Similarly, z 2i When (t) is 1, the high-efficiency path Lq3 is used as a constraint condition. That is, z 2i When (t) is 1, when the number of operating units is 2, the highest-efficiency operating point is used as a constraint. Similarly, z 3i When (t) is 1, the maximum output path Lq4 is used as a constraint condition. That is, z 3i When (t) is 1, when the number of operating units is 2, the maximum output operating point is used as a constraint.

[0065] Also, as shown in equation (10), when z 1i (t) is 1, by using equations (11) and (12) as linear constraint equations, the water consumption for power generation q i (t) and the power generation output p i (t) move on the high-efficiency path Lq1.

Number

Number

Number

[0066] Similarly, as shown in equation (13), when z 2i (t) is 1, by using equations (14) and (15) as linear constraint equations, the water consumption for power generation q i (t) and the power generation output p i (t) move on the high-efficiency path Lq3.

Number

Number

Number

[0067] Similarly, as shown in equation (13), z 2i When (t) is 1, by making equations (14) and (15) linear constraints, the amount of water used for power generation q on the high-efficiency path Lq3 i (t) and power output p i (t) moves.

number

number

number

[0068] The optimization processing unit 114 uses the relationships in equations (1) to (5) as constraints to perform an operation to maximize the evaluation function H shown in equation (6). In this case, the relationships in equations (7) to (18) are added as constraints. As described above, the optimization processing unit 114 performs an operation to maximize the evaluation function H, using 30 minutes (see Figure 11) as an example of the sampling Δt of the operation plan. Here, the optimization processing unit 114 can use, for example, a general solver corresponding to mixed-integer programming to calculate the solution to the optimization problem that maximizes H.

[0069] The planning generation unit 116 calculates the inflow amount q for each time step Δt of each dam Di (1 ≤ i ≤ n). ini (t), dam reservoir v i (t), total drop h i (t), amount of water used for power generation q i (t), discharge amount q di (t), power output p i (t) is represented in a chart. For this reason, when the plan generation unit 116 targets a two-day operation plan, it generates an operation plan that includes at least the information for nt=96, i.e., t=1, 2, ..., 96 (see Figure 11).

[0070] The control unit 104 displays the operation plan table for each power plant Ppi (1 ≤ i ≤ n), generated by the plan generation unit 116, on the display device 20. The control unit 104 can also transmit the plan to the power generation control device of each power plant Ppi (1 ≤ i ≤ n) for automatic operation. In water systems operated by operators, the power plant operators can also refer to the operation plan on the display device 20 and operate the water system themselves.

[0071] Figure 13 is a flowchart showing an example of processing by the hydroelectric power plant operation support device 10. As shown in Figure 13, first, the acquisition unit 100 stores the operating point, initial water level, and final water level of each power plant Ppi (1≦i≦n) selected by the operator in the operation characteristic data storage unit 14 via the input device 30 (step S200). Here, the operating point is the highest efficiency operating point and the power plant's maximum output operating point, and can be selected for each power plant Ppi (1≦i≦n).

[0072] Next, the acquisition unit 100 receives the inflow amount q during the forecast period via the network nw. ini (t)(1≦i≦n) and the current total drop hi(1≦i≦n) are obtained and stored in the driving characteristic data storage unit 14 (step S202).

[0073] Next, the evaluation function generation unit 110 generates an evaluation function based on this initial information (step S204). Subsequently, the optimization calculation processing unit 114 performs an operation to maximize the evaluation function H shown in equation (6) using the relationships in equations (1) to (5) (step S206). The constraints in this process are the relationships in equations (7) to (18).

[0074] Then, the planning generation unit 116 calculates the inflow amount q for each time step Δt of each dam Di (1≦i≦n). ini (t), dam reservoir v i (t), total drop h i (t), amount of water used for power generation q i (t), discharge amount q di (t), power output p i (t) is visualized as a graph and output to the display device 20 (step S208), and the process is terminated.

[0075] In this way, a water system operation planning method can be provided that enables maximum operation of power generation by selecting between high-efficiency operation, which includes the time change of the total head (static head) hi that changes in accordance with the change in the amount of water stored in the dam Di (1 ≤ i ≤ n), and pipeline pressure loss l of the branch pipelines which differs for each number of operating units N (see Figure 4), and maximum output operation of the power plant. This makes it possible to generate operation plans with higher accuracy.

[0076] Furthermore, for each dam Di (1≦i≦n), an operating point is selected from the high-efficiency operating point and the maximum output operating point, and these are added to the constraints to maximize the evaluation function. This eliminates the need to repeatedly derive the optimal solution by updating the constraints to match the head due to changes in dam water volume in the optimization process that maximizes the evaluation function. This reduces the computational load while suppressing the decrease in the accuracy of the hydroelectric power plant operation plan. Therefore, it becomes possible to consider the total head (static head) hi due to changes in dam water volume without a trade-off between computational load and planning accuracy, and a practical water system operation plan can be provided.

[0077] Furthermore, by calculating the water system operation plan within the operating points, including the highest efficiency operating point for each number of operating units N for each power plant Ppi (1≦i≦n), and the power plant's maximum output operating point, it becomes possible to suppress the number of changes in the operation of the water usage for each power plant Ppi (1≦i≦n) for self-generation water systems operated by operators. This supports efficient water system operation by operators. In addition, by selecting the combination of operating points to output as an operation plan from the operating points included in the operation characteristics data for each power plant Ppi (1≦i≦n), it is possible to provide a water system operation plan tailored to the needs of the power plant.

[0078] As explained above, according to this embodiment, the evaluation function generation unit 110 generates the time-series amount of water used for power generation q drawn from the water storage of the dam Di (1 ≤ i ≤ n). i The power output p changes in time series according to (t). i The sum of the power generation value for (t)(1≦i≦n) is generated as the evaluation function H((Equation (6))). Then, the optimization calculation processing unit 114 calculates the time-series power generation water usage qi (t)(1≦i≦n) and the time-series power output p i The relationship between (t)(1≦i≦n) and the water storage volume v of dam Di(1≦i≦n) is expressed as follows: i (t)(1≦i≦n) - drop h i (t) Constrained to a predetermined operating point that changes according to (Equation (5)) (Equations (7) to (18)), the time-series water usage q for power generation is set such that the evaluation function H is maximized. i We decided to generate planned values ​​for (t)(1≦i≦n).

[0079] This eliminates the need to repeatedly derive the optimal solution by updating constraints to match the drop in water level due to changes in dam water storage volume in the optimization process that maximizes the evaluation function H. This reduces the computational load while suppressing the decrease in accuracy of the power plant Ppi (1 ≤ i ≤ n) operation plan. [Explanation of symbols]

[0080] 1: Hydroelectric power plant operation support system, 10: Hydroelectric power plant operation support device, 20: Display device, 30: Input device, 40: Prediction device, 100: Acquisition unit, 102: Characteristic data generation unit, 110: Evaluation function generation unit, 114: Optimization calculation processing unit, 116: Plan generation unit, Di(1≦i≦n): Dam, h i (t)(1≦i≦n): Total head, l: Pipe loss length, Pi (1≦i≦n): Hydrogenerator, p i (t)(1≦i≦n): Power output (amount of power generated), Ppi(1≦i≦n): Power plant, qi(1≦i≦n): Water used for power generation, qini(1≦i≦n): Inflow, v i (t): Dam water storage volume

Claims

1. An evaluation function generation unit generates an evaluation function based on the power output of a generator that changes over time according to the time-series amount of water used for power generation drawn from the dam's reservoir, An optimization calculation processing unit generates a planned value for the time-series power generation water usage in order to maximize the evaluation function by constraining the relationship between the time-series power generation water usage and the time-series power generation output with respect to the time-series power generation water usage to a predetermined operating point that changes according to the head based on the water storage volume of the dam. A hydroelectric power plant operation support device equipped with [specific features / features].

2. The hydroelectric power plant operation support device according to claim 1, wherein the time-series water usage for power generation is a planned value corresponding to the time-series predicted value of the dam inflow into the dam, and the head changes in a time series based at least on the time-series predicted value of the dam inflow and the time-series planned value of the water usage for power generation.

3. The hydroelectric power plant operation support device according to claim 1, wherein the operating point is the operating point with the highest efficiency at which the power generation efficiency of the power output relative to the amount of water used for power generation is maximized.

4. The hydroelectric power plant operation support device according to claim 3, wherein the operating point is the operating point of the maximum output at which the power generation output is maximized relative to the amount of water used for power generation.

5. The hydroelectric power plant operation support device according to claim 4, wherein the optimization calculation processing unit can select either the operating point with the highest efficiency or the operating point with the highest output as the constraint.

6. The optimization calculation processing unit restricts the power generation output to an operating point corresponding to the number of turbine generators when the power generation output is the output of multiple turbine generators, as described in any one of claims 1 to 5, for the operation support device of a hydroelectric power plant.

7. An operation plan generation unit generates an operation plan that includes at least one of the time-series changing head, the time-series planned value of the amount of water used for power generation, and the time-series power output. The hydroelectric power plant operation support device according to claim 6 is further provided.

8. The aforementioned head is the total head, and by reducing the pressure loss in the pipeline used for power generation, it becomes the effective head. An operating characteristics data storage unit stores pipeline loss characteristic data showing the relationship between the amount of water used for power generation and the pipeline pressure loss, turbine characteristic data showing the turbine efficiency of the turbine generator which changes according to the effective head, the rotational speed of the turbine generator, and the amount of water used for power generation, and generator characteristic data showing the generator efficiency of the turbine generator which changes according to the power output of the turbine generator. A characteristic data generation unit generates information about the operating point using the pipeline loss characteristic data, the turbine characteristic data, and the generator characteristic data, and stores it in the operating characteristic data storage unit. Furthermore, The hydroelectric power plant operation support device according to claim 7, wherein the optimization calculation processing unit generates a planned value for the time-series power generation water usage using the operating point information stored in the operating characteristic data storage unit.

9. The hydroelectric power plant operation support device according to claim 8, wherein the optimization calculation processing unit can define the operating point as a linear form using the amount of water used for power generation and the power output.

10. The hydroelectric power plant operation support device according to claim 9, wherein the evaluation function is an added value assigned as a coefficient of the power output when maximizing the amount of power generated, and the unit price of electricity is an added value assigned as a coefficient of the power output when maximizing the amount of power sold.

11. The hydroelectric power plant operation support device according to claim 10, wherein, in the case of maximizing the electricity-to-water ratio, which is the ratio of the amount of electricity generated to the amount of water used for power generation, the operating point with the highest efficiency is used as a constraint condition.

12. An evaluation function generation process that generates an evaluation function based on the time-series change in the power output of a generator according to the time-series amount of water used for power generation drawn from the dam's reservoir, An optimization calculation process that generates a planned value for the time-series water usage for power generation such that the evaluation function is maximized by constraining the relationship between the time-series water usage for power generation and the time-series power output with respect to the time-series water usage to a predetermined operating point that changes according to the head based on the water storage volume of the dam, A method for supporting the operation of a hydroelectric power plant, which includes the following features.

Images