System and method for optimising and controlling hydroelectric power stations

Abstract

The purpose of the present invention is to make the best use of water resources, and the technology described uses data collected through sensors integrated into the hydraulic turbine which measure the dynamic differential pressure and the power in the electric generator, so as to feed a control system which allows calculation of an index which indicates the optimal efficiency of the turbine-generator set in real time, based on an adaptation of the Winter-Kennedy equation, without measuring flow or reduced model data or field efficiency measurements.

Description

“SYSTEM AND METHOD FOR OPTIMIZATION AND CONTROL OF HYDROELECTRIC POWER PLANTS” Field of Application

[0001] The technology is used in hydroelectric power generation, acting in the control and operation of the generation system, aiming to optimize the conversion of available hydraulic energy (head and flow) into electrical energy by monitoring all variables that determine the efficiency of the hydraulic turbine-generator set in real time. The main variables are flow and net head, which are altered over time by downstream and upstream level conditions, pressure drop in the hydraulic circuit, and mechanical and electrical wear on the turbine-generator set, among other factors. These conditions, which vary from moment to moment, make the point of best efficiency dynamic and must be continuously determined to optimize the operation of the turbine-generator set.Therefore, this technology addresses the concept of a data acquisition system and its correlation to determine the best operational efficiency point, without the need for reduced model data, flow measurements, or field efficiency tests. Prior Art.

[0002] Hydroelectric power is a clean, renewable energy source, among the most widely used worldwide, produced by the movement of water retained in a dam through a turbine-generator system. It is also one of the most important energy sources globally, according to data from the International Energy Agency. (IEA), in 2020, hydropower accounted for about 16% of total electricity production worldwide.

[0003] One of the main advantages of hydroelectric power is its predictability. Once a watercourse is present, it will always be in motion, ensuring a constant flow of energy, unlike intermittent energy sources such as solar and wind, which can suddenly cease generation. Furthermore, it does not require fossil fuels for its generation, significantly reducing its long-term environmental impact.

[0004] Water-generated energy is also a flexible energy source. It can be easily adjusted to balance energy and environmental demands at different times, simply by managing the reservoir, where appropriate. This means generation can be increased during periods of peak demand, when other energy sources may not be able to meet it economically.

[0005] There are some important challenges that need to be addressed to expand the supply of hydropower, such as the availability of suitable sites for new hydropower plants, as many of the best areas for hydropower have already been explored, and finding new sites with economically and environmentally viable hydropower potential can be difficult. This is in addition to social conflicts, as the construction of hydropower plants often involves the displacement of local communities.

[0006] Environmental impact is another challenge in building new hydroelectric plants, which can have a significant impact on the environment, including flooding and disruption of natural habitats. It is important to balance energy needs with environmental preservation.

[0007] The increase in global temperature is also a factor to be evaluated, as changes in rainfall patterns can affect the flow of water in rivers and reservoirs, reducing the availability of water to generate hydroelectric power.

[0008] Costs are also a challenge, as building new hydroelectric plants can be expensive and require intensive investment. Furthermore, maintaining and upgrading existing plants also requires regular investment.

[0009] Therefore, efficient management of available resources is crucial. Older hydroelectric plants can be modernized and / or expanded to increase efficiency and power generation capacity. This may include the installation of cutting-edge technology, such as more efficient turbine-generator sets and real-time automated control systems.

[0010] Figure 1 presents an estimate of turbine-generator set wear, prepared by EPE (Brazilian Energy Planning Company). This wear, in addition to other conditions, causes the optimum efficiency point to change over the plant's lifespan.

[0011] One of the consequences of the improvement of the energy generation and sales market is the increase in Competitiveness demands increasingly efficient processes and equipment from generators, extracting the maximum possible energy from natural resources with the least impact. This capacity becomes a crucial factor in selecting potential investments and even in granting and granting natural resources by the responsible agency (ANEEL).

[0012] Specifically, in the case of hydroelectric generation, the system depends directly and extensively on natural factors such as rainfall, ambient temperature, fluid density, etc., making it essential to calibrate the equipment correctly to ensure maximum resource utilization during operation. Traditionally, system calibration is performed by operating the turbine-generator set at the point of maximum theoretical efficiency determined during the design stages. This usually involves starting and stopping cycles to compensate for variations in flow, droop, and energy demand over time.

[0013] During the development of the hydraulic, mechanical, and electrical designs of the turbine-generator set, it is possible to define, calculate, measure, and certify the actual theoretical efficiency of the turbine-generator set through reduced-scale models tested in the laboratory. However, their high cost means they are mostly adopted in large hydroelectric plants, which can afford such an investment. In the case of small plants, previously tested models are adopted, complemented by sophisticated fluid mechanics calculations, the objective of which is to best adapt the hydraulic profile to the specific design parameters.

[0014] Despite being sophisticated, both models are insufficient, since they only approximate real conditions, making it impossible to predict all the interferences in the model, also resulting from the equipment manufacturing processes (turbine-generator sets, valves, penstocks, gates, and others) and the implementation of the plant's physical structure (adduction channel and tunnel, tailrace, upstream and downstream elevations, gross head, net head, and others). In plants with more than one generating unit, even with identical power output and mechanical and electrical designs, differences in efficiency are invariably observed between the generating sets due to variations in the conditions in the hydraulic circuit of each one and / or also due to their construction, especially with regard to the hydraulic profile of the blades and their mounting on the rotor.Other factors that contribute to deviations from ideal generation conditions are: the amount of water and head available and the requirement for continuous generation, cleaning of the grids, wear or cavitation action in the rotors, misalignment of the guide blades, silting of the tailrace channel, and others.

[0015] To compensate for the design limitations of the aforementioned alternatives, efficiency tests are commonly performed on field-installed equipment to validate design criteria. These tests determine the net head, turbine flow, and generator power—variables from which efficiencies can be calculated over a predetermined range of loads and hydraulic conditions. However, this process is generally costly, requires interruption of generation, and provides only a snapshot of the current operating condition at the time of the tests. tests, disregarding the inherent lag of natural variations in operation and wear of equipment over time.

[0016] The Winter-Kennedy method was developed to perform calibration tests and consists of the construction of a simple measurement system, which includes a precise differential dynamic pressure transducer, hydraulically connected to the pressure receptions in the spiral housing and, electrically, by the measurement module to the computerized data acquisition system.

[0017] It is one of the oldest and most widely used methods in hydroelectric installations worldwide. Major users of this method include Hydro Québec (Canada's largest power utility) and Brazil's CEMIG (a power utility in Minas Gerais), which is present at the Três Marias Hydroelectric Plant, built in 1960, and the Igarapava Hydroelectric Plant, built in 1999.

[0018] The method was developed by IA Winter and AM Kennedy in the 1920s; it is applied to flow measurement in a turbine-hydraulic generator set and is based on equating the flow rate and the static pressure difference between two pressure taps located at the inner radius of the spiral casing, close to the turbine rotor, and at the outer radius of the spiral casing, away from the rotor, as can be seen in Figure 2.

[0019] Flow rate is generally considered an important variable in the industrial environment. Besides being linked to the quality of the products to be produced, it is also linked to the cost of those products, the raw materials, raw materials and other products that will be added at the same stage of the process.

[0020] Flow rate is also used as a set point or target value that an automatic control system will attempt to achieve, increasing the operational efficiency of high-cost and critical industrial equipment, such as conventional power generators, combustion engines, turbogenerators, steam generators, etc. With the flow rate within the set point, these devices will operate satisfactorily, without losses, without errors, and within certified standards. This ensures the proper functioning of these devices and helps extend their useful life. If a generator's consumption flow rate is not in line with the ideal set point, it certainly has a calibration issue that needs to be corrected as soon as possible. The importance of flow rate as a process variable is infinite.

[0021] This method employs a factor that relates the flow rate to the pressure difference measured between these two points, determined in a scale model and then applied to the installation. It has the advantage of using low-cost equipment and can be applied to conduits of any diameter. One disadvantage is the need for a scale model. It is recommended for new projects, since building a scale model solely for flow measurement is impractical.

[0022] The method is recommended by the international standard IEC 41, especially for optimization tests of classical turbine-generator sets or Kaplan Bulb turbine-generator sets. It is assumed that the systematic uncertainty of this method is equal to the uncertainty of the measurement method used for calibration.

[0023] The Winter-Kennedy equation is obtained from equation 1: ^^ = ^^. ∆ ^^ n(1) Where: Q - flow rate (m³ / s); ^P - pressure difference between Winter-Kennedy points (Pa); K - constant obtained from Winter-Kennedy (ms -1 . Shovel- n ), n - exponent obtained during calibration of the method, which appears in the literature with values ​​equal to or close to 0.5.

[0024] The constant K, which relates the flow rate to the pressure difference between two points, must be determined in the reduced model test.

[0025] Regarding implementation feasibility, one limitation of the method is the need for a reduced-scale model, as the factor that relates flow to the pressure difference between two points must be determined in the reduced-scale model. The major advantage of this method over others is the use of low-cost installation equipment, which makes it widely used. Once installed without interference, the method's accuracy ranges from ±1 to 1.2% (Borges, RV; Vasconcelos, P. S; Ventura, LH G; and Mensah. JHR HYDROMECHANICAL COMPONENTS - COMPARING THE WINTER KENNEDY AND GIBSON FLOW MEASUREMENT METHODS).

[0026] Comparison of the performance of the same A turbine-generator set under various conditions can be analyzed using equations 2, 3, and 4 and verifying whether there are variations in flow and power values ​​due to, for example, different net heads. A net head of 30 m, when reduced by 1 m (-3.33%), reduces power by 4.95%, i.e., the power loss is 48% greater compared to the reduction in head. Federal University of Itajubá – MG Institute of Natural Resources – IRN

[0027] As can be seen, theoretically or practically, such analysis methods create a punctual and inefficient portrait of the operating conditions, requiring real-time analysis to ensure the optimized efficiency of the turbine-generator set.

[0028] Similarly, to optimize real-time operations and address these challenges, document ES2648988T3 describes real-time plant monitoring, comparing regular operation to operation with the optimization system. This document was generated from an index test, based on the comparison of data obtained through additional sensors installed and integrated into the plant structure with the models. pre-established computational models in the project to obtain operational efficiency indicators on a regular basis. It is important to emphasize that this test is limited to the equipment's behavior in operation, through the analysis of parameters such as mechanical vibrations, temperature, turbine-generator set rotation speed, and others. Despite several benefits, such as maintenance predictability and a reduction in unplanned shutdowns due to failures, this system does not visualize flow and head conditions in real time. It depends directly on computational models established during the plant's design, which do not generate precise gains in optimizing the conversion of hydraulic energy into mechanical / electrical energy in real time.

[0029] In contrast, the concept proposed by the technology being protected acquires relevant data with high reliability and precision, from which it analyzes and models optimal operating conditions, taking into account flow and head conditions at each instant. This data is obtained through differential dynamic pressure sensors installed in the hydraulic turbine and through power measurements in the electric power generator. Through various mathematical and computational models, these data are appropriately processed so that they can be used in optimization models for the conversion of hydraulic energy into electrical and mechanical energy in real time. Description of Figures

[0030] Figure 1 represents the effect of machine wear on power generation over time. over the years, the generation gains brought about by technological progress and the effects that restoration and repowering actions promote in a hydroelectric plant.

[0031] Figure 2 shows a block diagram of continuous flow measurement through the turbine / generator set, using the differential dynamic pressure method.

[0032] Figure 3 shows the optimum operating point of the turbine-generator set for a given condition.

[0033] Figure 4 indicates the optimum operating point of the turbine-generator set for different gross hydraulic heads. Description of the Patent Application

[0034] The proposed concept seeks to optimize the energy production process, dispensing with direct and continuous measurement of inflow and net drop of the turbine-generator set, or data obtained through tests on a reduced model, or even efficiency tests carried out in the field.

[0035] The new concept is based on the acquisition and processing of differential dynamic pressure data measured in the hydraulic turbine and the power in the electric generator. These measurements are obtained using different types of hardware and with adequate precision to obtain accurate values.

[0036] This equipment may be based on a central processing unit, such as a computer or programmable logic controller. This equipment will have access to the quantities required for algorithm processing. These quantities will come from the plant's local automation system, acquired by sensors such as the dynamic differential pressure transmitter located at the pressure taps on the turbine / generator assembly and the speed regulator.

[0037] Speed ​​governors (SRs) are an integral part of a plant's digital automation and control system. They are usually exclusive to each Generating Unit (GU) and regulate the torque on the turbine / generator shaft through hydraulic subsystems, controlling the opening of the distributor and rotor (in a Kaplan turbine / generator assembly). System control acts on the SRs by adjusting the active power setpoint (or distributor opening) via industrial protocol communication (e.g., Modbus, OPC-UA, EtherNet / IP). Only digital governors with an industrial protocol communication module are compatible with the optimization system.

[0038] From this concept it is possible to obtain a coefficient that is directly proportional to the real and instantaneous performance of the equipment, using a variation of the Winter-Kennedy equation added to the other variables obtained via communication in, for example, a controller designated for this function, according to equation (8).

[0039] This method allows, manually or automatically, to adjust the power through regulators present in the plant's control and automation panel, adapting its operation to the point of greatest performance, improving the use of the water resource used, increasing the energy generated according to the available water resource, as long as the set of other operational factors allows it (sanitary flow, environmental restrictions, dispatch power, etc.).

[0040] To define the new control coefficient, we start from the Winter-Kennedy equation defined by the constants ken, that is, which do not vary over a certain operating range, around the nominal power of the turbine / generator set: ^^ = ^^∆ ^^ ^^ (1)

[0041] The equation for the power generated by the turbine / generator set is: ^^ = ^^ ^^ ^^ ^^ℎ (5)

[0042] Where: P is the mechanical power on the shaft of the turbine / generator set; η is the efficiency; ρ is the density of the fluid; g is the acceleration of gravity; h is the net head.

[0043] Considering that ρ and g are constant at each instant of the measurement, equation (5) can be described as: ^^ = ^^ℎ ^^1^^2∆ ^^ ^^ (6)

[0044] Since h, k1 and k2 are constants for a given instant, we can rearrange equation (7) as follows: ^^ = ^^ ^^3∆ ^^ ^^ (7)

[0045] The relationship that optimizes generation is given by equation (8): ^^ ^^ ^^ = ^^3∆ ^^ ^^ or ^^ ∝ ∆ ^^ ^^ (8)

[0046] Therefore, to find the best generation condition (optimized generation) it is necessary to obtain The calculation input data, which are the differential dynamic pressure (^P) in the hydraulic turbine and the electrical power in the generator (P), are used. To achieve this, the calculation system will utilize a new component called the "Main Controller" that will be inserted into the existing automation network at the generation unit. This component is responsible for acquiring and processing the signal from the differential dynamic pressure transmitters and executing the performance optimization control algorithm. This algorithm combines multiple functions, acting as a gateway for converting communication protocols, logging data locally, and communicating with optional cloud services. The connection is protected by security algorithms on both ends.

[0047] Optionally, the results obtained by the performance optimization system can be made available through specific applications via the internet in real time, as well as stored for later analysis.

[0048] When starting a program, for example, this device enters an "Idle" state until it receives a command enabling one of the control modes. Except for the scan routine, which is finite, programs enter a cyclic execution mode, being disabled only if the protection is triggered ("Trip").

[0049] Since the efficiency coefficient scale is not absolute, the control method must be discrete. To this end, the P&O (Perturb and Observe) algorithm was adopted in this example. This algorithm can be optimized using Artificial Intelligence techniques such as Fuzzy Logic. The control premise is based on finding the point of best performance through the introduction of sequential disturbances that reduce over time, comparing the current and previous states until, finally, convergence on the optimal operating point occurs.

[0050] The system also allows for plotting the performance curve for the machine's instantaneous operating condition, which is not exactly the same as the nominal design conditions, even for two mechanically identical machines. The scan routine performs tests in a predetermined power range with a configurable step and stores the data in memory or makes it available via communication. Plotting the data yields results similar to the output graphs of a standard index test.

[0051] This technology can be better understood with the following non-limiting examples. Example 1: Determining the optimum operating point of the turbine-generator set The determination of the highest value for the calculated index that identifies the optimum operating point of the turbine-generator set for a given condition can be seen in Figure 3.

[0052] The determination of the highest value for the calculated index that identifies the optimal operating point of the turbine-generator set for different gross hydraulic heads (for example: 30m x 29m), can be seen in Figure 4, where it is possible to observe that the optimal power for the two head situations considered is different: from 4,050 kW to 3,900 kW.

Claims

1 / 2 CLAIMS 1. Hydroelectric power station optimization and control system that dynamically optimizes turbine efficiency, consisting of main components inserted into the existing automation network of the generation unit. The central processing unit acquires and processes the signal from dynamic differential pressure transmitters through the acquisition of sensor signals and executes the performance optimization control algorithm. This unit processes and converts protocols and makes them available for consumption on a cloud platform, serving as data storage and replacement in cases of connectivity loss; in addition to governors, digital speed regulators that regulate the torque on the shaft of said turbines. 2.Hydroelectric power station optimization and control system according to claim 1, characterized in that the central processing unit operates through a logical architecture integrated with the existing control systems in the plant, such as speed governors, sensors and automation controllers.

3. Hydroelectric power station optimization and control system according to claims 1 and 2, characterized in that a central processing unit operates through a finite state machine in which, once the program is started, said machine enters an idle state until it receives a command enabling one of the control modes, except for the case of the scan routine, which is infinite; the programs enter a cyclic execution mode and. 2 / 2 are disabled only in case of protection actuation (“Trip”), in which case a reset signal can be triggered, returning to the standby state.

4. Hydroelectric station optimization and control system, according to claims 1 to 3, characterized in that the control algorithm performs tests in a predetermined power range, with a configurable step, which stores the data in memory or makes it available via communication, and said algorithm finds the point of best performance by introducing sequential disturbances, which reduce based on time and the comparison between the current and previous states until, finally, convergence on the optimal operating point occurs, allowing the plotting of the performance curve for the instantaneous operating condition of a turbine, even if such conditions deviate from the nominal design conditions. 5.Method for optimizing the control of hydroelectric stations, characterized by occurring through the relationship between an efficiency coefficient and the ratio between the power generated and the differential pressure in the turbine spiral casing raised to the Winter-Kennedy exponent.

Images