GB2405492A

GB2405492A - System for remote monitoring and control of power generating plant

Info

Publication number: GB2405492A
Application number: GB0320406A
Authority: GB
Inventors: Yunfei Bai
Original assignee: DISTANT CONTROL Ltd
Current assignee: DISTANT CONTROL Ltd
Priority date: 2003-08-30
Filing date: 2003-08-30
Publication date: 2005-03-02
Anticipated expiration: 2023-08-30
Also published as: GB0320406D0; GB2405492B

Abstract

A system for remote monitoring and control of at least one power generation plant 8, 11 comprises a first computing means 10 communicating with the power generating plant 8, 11 and a second computing means 1 remote from the first computing means. The first computing means 10 is operable to receive information from and deliver control to the power generating plant 8, 11. The second computing means 1 is operable to receive data from the first computing means 10, manipulate the data and apply a performance model of the power generating plant 8, 11 to generate control commands. The first and second computers 10, 1 may be connected by internet or wireless network 14. The performance model may be a characteristics and performance model.

Description

Method and Apparatus for Remote Control of Power Plants The present

invention relates to apparatus and a method for the remote monitoring and control for power generation S systems.

Renewable energy sources have attracted public attentions in recent years.

A typical renewable energy power system consists of a power source, a balance system and a power load. The power source can be solar array, wind turbine, and hybrid system of solar and wind power sources etc. The balance system may include DC-DC and/or DC-AC converter/inverter and drive circuits for power conversion and conditioning. The load of the system can be the utility grid or any DC or AC loads.

Most of the existing systems either use a conventional control strategy, or use an inaccurate mathematical model to do the power delivery. Both of these two methods can cause more power loss and hence reduce the energy conversion efficiency.

References herein remote monitoring, and control should be taken to include intelligent modelling (using AI methodologies), control, monitoring, analysis and diagnose, and administration of remote power plants It is an object of the present invention to address the above mentioned disadvantages.

According to a first aspect of the invention a system for remote monitoring and control of at least one power generation plant comprises: first computing means communicating with said at least one power generation plant, said first computing means being operable to receive information from and deliver control to the at least one power generation plant) and second computing means operable to receive information from the first computing means and to manipulate said data to apply a performance model of the at least one power generation plant, and generate control commands based on said model) wherein the first and second computing means are remote from each other and are operable to communicate over a network of interconnected computing means.

The performance model is preferably a characteristics and performance model.

The application of a characteristics and performance model is preferably to achieve one or more of the following: to demonstrate the system operation, to apply a characteristics and performance model of the at least one power generation plant, analysis system performance, detect potential faults and generate control commands.

The network of interconnected computers may be computers, communication devices, personal digital assistants (PDAs), etc., that allow remote operation from one device to another by means of for example Internet, telecommunication network or wireless network such as Blue-tooth and/or the 802.lla/b/g standard (Wi-Fi).

Preferably, the at least one power plant is a renewable energy power plant, which preferably incorporates a power source and a balance system. The power plant preferably communicates with a power load, which may be a powered device or a power grid/power supply network. Preferably, the system comprises first computing means for each of a plurality of power generation plants.

The first computing means preferably comprises a power plant data sampling element, preferably operable to receive data from sensors of the power plant and/or from the balance system and/or from environmental sensors in the vicinity of the power plant.

The first computing means preferably incorporates a control execution element; preferably operable to receive commands from the second computing means and to deliver control to the power generation plant, in particular the balance system and power source. The first computing means may also include a network interface element to enable different types of network connection from local area network connection, Internet connection to wireless network connection.

The second computing means preferably includes an application module. The application module is preferably operable to communicate with the first computing means.

The application module is preferably operable to communicate with third computing means which may be one or more of a remote computer, remote communications device and/or a remote personal digital assistant (PDA), which third computing device(s) is/are preferably remote from both the first and second computing means.

The second computing means preferably includes a modelling element. The modelling element is preferably operable to model data provided by the first computing means from the power generation plant and/or associated sensors. Said modelling element may use neural network algorithms, fuzzy logic, evolutionary computation, expert systems and/or optimization algorithms to generate models.

The second computing means preferably includes a control element, which provides control signals to the first computing means based on the output of control algorithms, such as optimal control, adaptive control, Proportional- Integral-Derivative (PID) control, fuzzy logic control and/or expert system control algorithms. The control element is preferably a self- learning element, such as a neural network element operable to adjust control command output based on previous behaviour of the power plant, or environmental factors affecting the power plant. The control element may alternatively be downloaded from the second computing means and executed in the first computing means.

< The second computing means preferably includes a fault/performance element, which is preferably operable to -- 30 receive data from the or each first computing means for the diagnosis and/or detection of faults. Preferably, the fault/performance element is operable to provide commands to the or each computing means to address and/or correct detected faults/errors. The error correction commands can also be provided remotely using a computer, remote communication device and/or PDA via network.

The second computing means may also receive data from the or each first computing means and may demonstrate an operation status for remote monitoring over the network.

The second computing means preferably includes a database module, which preferably includes models of performance/behaviour of environmental factors affecting the power generation plant and/or the power source and/or the balance system.

The database module may incorporate control scheme data for providing to the control element.

The database module may incorporate a data element for storing information received from the first computing means. The database module may also incorporate a user's information data, power plant information data and/or other relevant data for system security.

According to a second aspect of the invention a method of controlling and monitoring at least one power generation plant comprises: receiving status information from the at least one power plant and passing control signals to the at least one power plant with first computing means; said control signals being received from second computing means, which second computing means also receive said status information and apply a performance model of the at least one power generation plant to generate controlling signals to be passed to the first computing means to be applied to the power generation plant) wherein the first and second computing means are remote from one another.

The performance model is preferably a characteristics and performance model.

The method includes the second computing means modelling power plant performance based on data received from the first computing means.

The invention extends to a power generation system having a remote control/monitoring system as defined in the first aspect. Preferably, the power generation system incorporates a power generation plant and the first and second computing means of the first aspect.

The invention extends to a computer program product operable to perform the method of the second aspect.

The computer program product may be in executable two parts, with a first part on the first computing means and the second part on the second computing means.

The invention extends to a computer programmed to perform the functions of the second computing means described in the above aspects.

Preferred features of the system are as follows.

A universal, open standard, web-based intelligent system for remote metering and monitoring, intelligent modelling, analysis, control and administration for power plants (in particular those with renewable energy resources) through the Internet.

The renewable energy resources of the power plants may include but are not restricted to the solar power modules/arrays and wind turbines, and the hybrid system containing both of them. The applications above include but are not restricted to the standalone, remote home, off-grid and gridconnected applications. The system preferably consists of three sites physically being in different locations but logically interconnected via the Internet. They are the multiple plant sites, the intelligent server site and the remote client site. At the intelligent server site, an intelligent power application server and a database preferably operate to implement the functions of remote metering and monitoring, intelligent modelling, analysis, control and administration. At the plant sites, a web-enabled site apparatus is preferably developed and facilitated to collect environmental data and system output data, to communicate with the remote intelligent power application server, and to execute commands sent from server site. At the remote client site, the system is preferably accessible by administrators and end users at anytime and anywhere via computer and wireless-Internet terminals. The wireless- Internet terminals include but not restricted to the mobile phone and PDA.

The web-based site apparatus is preferably a micro- processor based embedded system consisting of a micro- processor, a data I/O module, a memory module containing data segments and programs, and an Internet interface module implementing different types of Internet connection. The data I/O module preferably consists of the input module for collecting samples from environmental sensors and power systems. The environmental measurements include but not restricted to the sunlight illumination, the ambient temperature, the humidity level, the wind speed, etc. Data of the power system may include but not restricted to the DC bus and AC bus voltages and currents, power and energy outputs, etc. The commands output from the data I/O module preferably include but are not restricted to signals to DC-DC converters, battery charge controllers and/or DC-AC PWM converters/inverters. The Internet connection module can preferably be either encapsulated in the apparatus, or externally connected to the apparatus via serial port. The connection may include but is not restricted to the 56 Kbps dial-up connection, lO/lOO Base-T connection and all types of broadband connection.

Different modelling techniques for power plants using artificial intelligence methodologies, including but not restricted to neural network, fuzzy logic, evolutionary computation, expert system, etc. The intelligent models preferably include but are not restricted to the current- power characteristic model for photovoltaic modules, maximum power point model for photovoltaic modules, the optimal Time-Speed-Ratio (TSR) model for wind turbines, the wind speed time series model, the wind power output prediction model, the turbine optimal speed vs. wind speed model, etc. The system is preferably designed as a generic open system using the advanced technologies, including but not restricted to the J2EE solution, .NET solution, etc. which is easy to extend and integrated with legacy power systems and other information systems. New versions of the site apparatus software are downloadable and hence the upgrade of the system is through Internet. The system is preferably designed as with four-tier architecture. They are the client tier, web tier, intelligent application tier, EIS (Enterprise Information System)/Database tier.

The modules operating in the intelligent module layer preferably implement the remote metering and monitoring, remote intelligent modelling, remote model-based analysis, fault detection and error correction, remote control and remote administration. The remote control strategies may include but are not restricted to model-based control, PID control, optimal control, adaptive control, fuzzy logic control, expert system control, neural network control, etc. All of the features described herein may be combined with any of the above aspects, in any combination.

The advanced artificial intelligence based methodologies proposed herein can solve the problems above by building up different types of intelligent models for renewable energy power generation systems For a better understanding of the invention and to show how the same may be brought into effect, specific embodiments will now be described by way of example and with reference to the accompanying drawings in which: Figure l is a simplified schematic representation of a system showing different types of renewable energy power plants, an intelligent power application server, a database and remote web clients interconnecting through the Internet.

Figure 2 is a schematic block diagram of the system showing connections between site apparatus and each component of the power plants at plant sites, as well as the functional modules of the intelligent power application server and the database located remotely to the plants.

Figure 3 is a schematic block diagram of web-enabled site apparatus consisting of processor, memory, data I/O and Internet interface module, and the relations to a power system and an Internet port.

Figure 4 is a schematic block diagram similar to Figure 3 showing an alternative site apparatus in which an Internet interface block is not embedded in the apparatus, but linked with other Internet connection equipment via serial ports.

Figure 5 is a schematic block diagram of a four-tier software architecture showing program and application modules running in each tier.

Figure 6 is a flow chart of an operation procedure between the site apparatus and the remote intelligent power application server.

To maximise the energy conversion efficiency of the balance system and reduce the power loss, a precise model is required for analysing the performance of the power station, diagnosing errors occurring during operation, forecasting the achievable power output, and providing control commands for the optimal inverter/battery charging controller. Using the model-based approaches the energy conversion efficiency can be significantly improved compared to the existing power plants without these models.

However, the characteristics of the renewable energy sources are usually influenced by the changing atmospheric conditions such as solar illumination, wind speed, ambient temperature, air humidity level, etc. In addition, solar panels and wind turbines often have complicated physical and electrical characteristics, which are difficult to represent explicitly by mathematical models. Furthermore, solar panels and wind turbines installed at different locations usually have different ageing characteristics.

For example, a solar panel working in tropical areas or humid areas may have more ageing than those that are in the dry and cool areas. All these factors make it of great importance to develop and implement a precise model of the power source in order to generate maximum power conversion efficiency in the balance system.

The apparatus (converter, inverter, charge controller, etc.) to implement the above mentioned intelligent strategies usually requires complicated arithmetic reasoning and computation, and hence requires high quality and high cost computer and electrical resources. This high cost makes them not applicable to the renewable energy market, and especially not viable for domestic use.

The solar panel itself may need little maintenance work, but the wind turbine and the balance system need to be checked often. Frequent fault detection and error correction are useful to ensure the system is running in good condition and to avoid unnecessary power loss.

However, the solar and wind power generation systems for remote applications such as telecom stations and ocean oil wells are located in remote areas. For these applications it is difficult to access those power systems to perform the ongoing maintenance. Also, it may be expensive, if not difficult, to frequently visit power plants located in different places to do maintenance work.

The present invention provides a complete Internet-based solution (or any other suitable network of interconnected computers that allow remote operation from one computer to another) to reduce the cost of the balance system, and solve the difficulties of maintenance of the renewable energy power plants. Using this solution, the intelligent strategy can be implemented by the cooperation between a world wide web-enabled site apparatus functioning locally at the plant site and an intelligent power application server remote to the plants. The site apparatus plays a role of communicator and executor. It collects data from the power system and sends them to the server, and receives server commands and executes on the balance system. All the intelligent strategies are performed in the remote server, and the apparatus applied at the plant site is therefore simple and with low cost. The server applies the multi-thread mechanism so that it is able to operate many plants located in different places, so the cost can be distributed and thus is not unacceptably high.

In addition, a portion of the maintenance work, such as operation monitoring, fault detection and error correction, is deliverable through the Internet. Plant site software can also be upgrade by downloading from remote server through Internet. Therefore the total cost of the balance system and the maintenance work can be IS significantly reduced.

The invention provides a universal, open-standard, web- based intelligent system with a four-tier architecture to perform the remote intelligent modelling, control, supervision and analysis to the renewable energy power generation systems.

Artificial Intelligence methodologies, such as neural network (see, e.g., Hassoun M H (1995) Fundamentals of Artificial Neural Networks. The MIT Press, 1995), fuzzy logic (see, e.g., Zadeh L and Kacprzyk J (1992) Fuzzy Logic for the management of Uncertainty. John Wiley & Sons, 1992), evolutionary computation (see, e.g., Baeck T et. al. (2000) Evolutionary computation 1: Basic Algorithms and Operators. Institute of Physics Pub, 2000), expert systems (see, e.g. Giarratano J C (1998) Expert Systems: Principle and Programming. Brooks Coles, 1998), etc. are applied to remotely establish different types of intelligent models for power plants. The models may include the power voltage model, the maximum power point tracking (MPPT) model, a wind turbine TSR model, a wind speed forecast model, a wind power prediction model, etc. Based upon these models, different type of intelligent control strategies and intelligent analyser are designed and developed to provide intelligent control to improve the power conversion efficiency, to evaluate the operating performance of the power plants, and to diagnose the potential system operation problems. In this way the energy conversion efficiency can be improved and the unnecessary power loss can be reduced.

Examples of intelligent modelling and control of solar array and wind turbines include the following.

Maximum Power Point (MPP) models for solar module/arrays can be built up using AI technologies. The aim of the model is to predict the maximum power point (the maximum voltage and current that can be generated) under varying weather conditions. Using neural networks, the input vector of the network can be set as weather conditions (radiation, ambient temperature, wind speed, humidity level, etc.) and the output is the predictable MPP. Sample data can be collected in real- time from environmental sensors and system outputs, and supplied to the model.

Different types of neural network architectures and training algorithms can be applied to generate the model.

Using fuzzy logic techniques, the relation between weather conditions and MPP can be identified and stored in heuristic fuzzy rules. The MPP can then be predicted by fuzzy reasoning mechanism based upon the online collected weather conditions. Using evolutionary computation technology, the extensive MPP candidates can be selected and evolved by different types of evolutionary computation algorithms, the optimal MPP can then be searched with respect to varying weather conditions. s

Time-Speed-Ratio (TSR)/Maximum Power Point (MPP) models for wind turbine are also able to predict the maximum mechanical power of the turbine corresponding to wind speed and blade pitch angle. Similar to the MPP modelling for solar arrays, the TSR of wind turbine with respect to the measurement of wind speed and blade pitch angle can be self-learned using online collected samples by AI modules.

These modules include but are not restricted to neural network, fuzzy logic, and evolutionary computation. The obtained TSR value can be used to calculate the power generator shaft speed and turbine torque, and hence MPP at any time.

When the models are obtained, the MPP generated by the models can be delivered to the control system as commands.

Using these commands, different types of control strategies can be applied to make the power generator operate in optimal condition, hence the maximum power delivery can be achieved.

Other models of the renewable energy generator can include but are not restricted to the solar array current-power characteristic model, wind speed time series model, wind power output prediction model, turbine speed - wind speed model, etc. The same AI principles can be applied to establish the models. These models can be used for system optimal control, system supervision and diagnosis, system performance and power delivery analysis, etc. Examples of ways in which efficiency can be improved are as follows. The optimal operation point (known as the maximum power point) of solar modules under a specific sunlight illumination and ambient temperature can be predicted by the AI-based model, and the control actions can be taken to make the power system work at the highest efficiency.

The optimal time-speed-ratio (known as TSR) corresponding to a specific wind speed can be predicted by the AI-based intelligent TSR model, therefore the maximum power coefficient of the wind turbine can be achieved by control the turbine shaft speed.

The optimal power coefficient of a wind turbine can also be achieved by adjusting the blade pitch angle based upon the intelligent model of the wind turbine mechanical power output and the winding current.

The intelligent but complicated schemes can be implemented by the cooperation between web-enabled site apparatus and a remote intelligent power centre over the Internet.

Logically the operation can be separated into two parts: the complicated algorithms are executed remotely from the power plant at the server site, and the results are sent to the power plant site apparatus as commands. Such commands may be optimal operating point of solar modules, duty ratio command and PWM modulating index and/or PWM waveform for DC-DC and DC-AC converters. The apparatus then delivers the commands to the power systems. Therefore the balance system comprising power inverter/converter/conditioner with a high performance can be simplified, thus the cost/benefit ratio can be significantly increased.

On top of this web-based architecture, the operation of the power plant can be supervised and manipulated in real- time at anytime and anywhere, and remote fault detection and simple error corrections can be delivered using computer or wireless-Internet terminals. An intelligent distributed control program can be downloaded to the local site apparatus and tuned at anytime and anywhere through the Internet. Therefore maintenance costs can be therefore reduced.

The system is designed as a generic open system using the advanced technologies (including but not restricted to the J2EE solution, .NET solution, etc.) with four-tier architecture referred to above which has features of good scalability and extensibility. The open standard utilised enables the system to be extended and/or integrated with legacy power systems and other information systems, such as MIS(Management Information System), ERP (Enterprise Resource Planning), CRM (Customer Relation Management), SCM (Supply Chain Management), B2B (Business to Business) and B2C (Business to Consumer), and other IT systems if necessary) with less cost and effort. In addition, new versions of the site apparatus software are downloadable and hence upgrading of the system is through the Internet.

Therefore the scalability, stability and reliability of the power generation system can be further improved.

The system is designed for the small scale and large scale renewable energy generation systems, including but not restricted to the solar power, wind power and hybrid systems. The applications of the presented invention include but not restricted to the standalone remote home, offgrid and grid-connected applications.

The system operates by the cooperation of the web-enabled site apparatus installed with power plants located in different places and the intelligent power application server and database located remote to the power plants.

Referring to Figure 1, the system consists of three sites, the multiple plant sites 8/11, 9, 10, 12, 13, the intelligent power application server site 1, 2, 3 and the remote client site 4/5/6/7. They are interconnected with each other via the Internet 14.

At the multiple plant sites, the renewable energy power sources, including but not restricted to solar array 8 and wind turbine 11, are connected to the local load 13 and/or utility grid 12 through the power converter/inverter 9.

Sensors and transducers 15 are facilitated to measure the environmental conditions (illumination, temperature, humidity level, wind speed) and system data (DC voltages and currents, AC voltages and currents, etc.). The web- enabled site apparatus 10 is used on each site connecting to the Internet 14 for data exchange, and to the inverter 9 for power conversion and generation control.

At the server site, the intelligent power application server 1 in which all the functional modules are implemented is connected to the Internet 14 through the firewall 16. A database system 2 is running together with the server 1.

The functional modules running on the intelligent power application server 1 are further depicted in the Figure 2 block diagram, which also shows a number of elements described in Figure 1; the same reference numerals apply.

It may include but is not restricted to the system administration module 19, the intelligent supervision and analyser module 20, the intelligent models 21 and/or the intelligent control strategies 22 for power plants. They communicate with the remote site apparatus through an Internet connection interface 18.

The remote client side consists of computers (see Figure 1) located either locally 3 or remotely 4 to the server site. In addition laptops 5, mobile phones 6, Personal Digital Assistants (PDAs) 7 (the latter two by a telecommunications network 17) and other wireless Internet terminals can be used to access the server 1 for remote manipulations.

Artificial Intelligence based modelling and analysis techniques Different types of intelligent models are built for the remotely operating power plants using artificial intelligence methodologies (see, e.g. Cawsey A (1998) The essence of Artificial Intelligence. Prentice Hall PTR, 1998). Environmental and system output data used for modelling are sampled using the site apparatus 10 and transmitted to the intelligent power application server 1 via the Internet 14. The environmental measurements include butis not restricted to the sunlight illumination, the ambient temperature, the humidity level, the wind speed, etc. Data of the power system may include but is not restricted to the DC bus and AC bus voltages and currents, power and energy outputs, etc. The AI methodologies, including but not restricted to neural network, fuzzy logic, expert system, evolutionary computation, etc. are used to build the intelligent models. The main features of these intelligent methods are that they have the capability of self-learning and allow representation of complicated characteristics that are usually difficult to be modelled and represented by explicit expressions, thus are suitable to represent the complicated physical and electrical characteristics of the renewable energy sources and systems. The intelligent models may include but not restricted to the MPPT model for solar arrays, the time- speed-ratio model for wind turbine, the wind speed forecast model, the wind power prediction model for wind turbine, etc. The web-enabled si te appara tus The web-enabled site apparatus is a micro-processor based embedded system. Referring to Figure 3 the system consists of micro- processor 26, memory 23 containing sections for plant data sampling, control execution, programs and data, analogue and digital I/O 24 for communication with the power converters and measurement instruments, and Internet interface module 25 enabling Internet data exchange by dial-up, Ethernet, broadband, or wireless network (such as Wi-Fi and Bluetooth) connection. These components are interconnected through the internal bus 27.

Referring to Figure 4, an alternative to Figure 3 is shown in which the Internet interface module 25 can be external to the site apparatus, such as a modem for Internet connection through a serial port 28.

The software system The software system has a four-tier web-based architecture, shown in Figure 5.

- Tier l The software running in the site apparatus lO (in Figure l) is the top tier. In addition, the web browsers running in computer 4, mobile phone 6, PDA 7 and any other Internet-enabled wireless terminals, used to access the intelligent power application server l, are also included in the top tier.

- Tier 2 A set of web components in the web tier, including but not restricted to HTML/XHTML, Java servlet, Java application, JSP (Java Server Page), ASP (Active Server Page), XML (Extensible Markup Language), WML (Wireless Markup Language), Java Script, VB Script, etc. forms the second tier. These components communicate with the on site programs and web browsers to receive requests and deliver responses. The results are presented in HTML/XHTML, XML, Java apples, Java bean, JSP, ASP, WML and other forms delivering to the computer 3, 4, 5 and wireless terminals 6, 7 at the top tier.

- Tier 3 What behind is the intelligent modules 30 tier, may be implemented by EJB (Enterprise Java Bean), CORBA (Common Object Request Broker Architecture), ActiveX, DCOM (Distributed Component Object Model), etc.., in which all the multi- thread intelligent modelling, control, analysing and diagnosing modules are facilitated. Receiving the requests from the web tier, all the functions are performed dynamically in this tier and the results are sent back to the second tier for presentation.

i) Remote metering and monitoring Basic information such as environmental measurement, power generation and consumption, is continuously collected and processed in the third tier. The results are organised and sent back as remote metering and remote monitoring. The remote metering and monitoring modules in this system enable users to get meter readings, monitor the power plants, and obtain the power generation and consumption at anytime and anywhere using computer 4/5, mobile phone 6, PDA 7, etc. The data is sampled from the plant site, analysed in the server and presented to the users via the Internet.

) Remote intelligent modelling The intelligent modelling processes are based upon the remotely collected samples from the plant sites. The models for power plants running in this tier may include but are not restricted to the power-voltage model, the MPPT model for solar arrays, the time-speed-ratio model for wind turbine, the wind speed forecast model, the wind power prediction model for wind turbine, etc. iii) Remote (model-based) analyses and supervision The intelligent models are used to further analyse the system operation performance. Based on the system output data consistently collected from the plants, potential faults (including but not restricted to short circuits, I output current distortion, large control errors, network congestions, etc.) can be detected, and non-optimal operating status can be perceived. Automatic system restarting and parameters resetting is then activated upon errors, meanwhile alerts are sent to a user's computer 6/7 and/or mobile phone 6.

iv) Remote control Intelligent control algorithms, which may apply the trained intelligent models, can be calculated and decisions are made at the intelligent power application server site. The intelligent control strategies may include, but are not restricted to PID control, optimal control, adaptive control, and other intelligent control strategies such as neural network control, expert system control, fuzzy logic control, etc. Commands can be sent to the site apparatus for execution. Programs and parameters can be downloaded to the plant site apparatus and control parameters can be adjusted and reset online in real-time via the Internet.

v) Remote system administration The whole system can be managed and administrated by system administrators at anytime and anywhere. The plant system can be started, stopped, paused and restarted remotely. The plant historical data can be manipulated, the on-site software can be downloaded and upgraded, and the intelligent agents and control schemes can be maintained through the Internet.

- Tier 4 The bottom tier comprises a large-scale distributed database archiving all the relevant data and program modules, including the intelligent models, control schemes, analysing strategy, and data sets collected from the remote plants.

The data communication between the intelligent application tier and the database may be through SQL, ODBC (Open Database Connectivity), Jdbc (Java database connectivity, ADO (ActiveX data object) etc. The environmental and system data are collected by the site apparatus, and sent to the remote intelligent power application server. As the data is received they are pre processed and then archived in the database as process data sets. The data can then be accessed by the web agents in the web tier and presented as a remote meter and monitor screen in the browsers of the client tier.

The data sets collected from different sites can be used to carry out the intelligent modelling for different types of power plants, both in online and offline manner, which are then archived as a TSR model, wind power model, MPPT model etc. in the database tier. Based upon these models and the predetermined diagnosis rules, the real-time operation process of the plants can be analysed. The potential faults can be detected by the diagnosis and fault detection module, the errors can be corrected by the error correction module, and the global performance can be analysed by the performance analyser module running in the intelligent application tier.

Furthermore, the completed models can also be used to generate control commands sent to the site apparatus for execution. The different types of control schemes are archived in the database tier, manipulated in the intelligent application tier, and downloadable to the site apparatus (client tier). The parameters can also be adjusted online.

In addition, the user information, plant information, and other relevant data are stored in the database tier for remote manipulation.

Figure 6 shows the collaborative operations between the web-enabled site apparatus and the intelligent power application server to implement the remote metering, modelling, analyser, supervision and control.

Claims

( CLAIMS: 1. A system for remote monitoring and control of at least one

power generation plant comprises: first computing means communicating with said at least one power generation plant, said first computing means being operable to receive information from and deliver control to the at least one power generation plant) and second computing means operable to receive information from the first computing means and to manipulate said data to apply a performance model of the at least one power generation plant, and generate control commands based on said model) wherein the first and second computing means are remote from each other and are operable to communicate over a network of interconnected computing means.
2. A system as claimed in claim l, in which the performance model is a characteristics and performance model.
3. A system as claimed in claim l or claim 2, in which the network of interconnected computers is computers, communication devices, personal digital assistants (PDAs), etc., that allow remote operation from one device to another by means of for example Internet, telecommunication network or wireless network such as Blue-tooth and/or the 802.lla/b/g standard (Wi-Fi).
4. A system as claimed in any preceding claim, in which, the at least one power plant is a renewable energy power plant.
5. A system as claimed in claim 4, which incorporates a power source and a balance system.
6. A system as claimed in any preceding claim, in which the system comprises first computing means for each of a plurality of power generation plants.
7. A system as claimed in any preceding claim, in which the first computing means comprises a power plant data sampling element, operable to receive data from sensors of the power plant and/or from the balance system and/or from environmental sensors in the vicinity of the power plant.
8. A system as claimed in any preceding claim, in which the first computing means incorporates a control execution element, operable to receive commands from the second computing means and to deliver control to the power generation plant.
9. A system as claimed in any preceding claim, in which the first computing means includes a network interface element to enable different types of network connection from local area network connection, Internet connection to
lO. A system as claimed in any preceding claim, in which the second computing means includes an application module, operable to communicate with the first computing means.
11. A system as claimed in claim 10, in which the application module is operable to communicate with third computing means being one or more of a remote computer, remote communications device and/or a remote personal digital assistant (PDA), which third computing device(s) is/are remote from both the first and second computing means.
12. A system as claimed in any preceding claim, in which the second computing means includes a modelling element.
13. A system as claimed in claim 12, in which the modelling element is operable to model data provided by the first computing means from the power generation plant and/or associated sensors.
14. A system as claimed in any preceding claim, in which the second computing means includes a control element, which provides control signals to the first computing means based on the output of control algorithms, such as optimal control, adaptive control, Proportional-IntegralDerivative (PID) control, fuzzy logic control and/or expert system control algorithms.
15. A system as claimed in any preceding claim, in which the second computing means includes a fault/performance element, which is operable to receive data from the or each first computing means for the diagnosis and/or detection of faults.
16. A system as claimed in any preceding claim, in which the second computing means is operable to receive data / from the or each first computing means and demonstrate an operation status for remote monitoring over the network.
17. A system as claimed in any preceding claim, in which the second computing means includes a database module, which includes models of performance/behaviour of environmental factors affecting the power generation plant and/or the power source and/or the balance system.

to
18. A method of controlling and monitoring at least one power generation plant comprises: receiving status information from the at least one power plant and passing control signals to the at least one power plant with first computing means; said control signals being received from second computing means, which second computing means also receive said it, status information and apply a performance model of the at least one power generation plant to generate controlling signals to be passed to the first computing means to be applied to the power generation plant; wherein the first and second computing means are remote from one another.
19. A power generation system having a remote control/monitoring system as defined in any one of claims 1 to 17.
20. A computer program product operable to perform the method of claim 18.
21. A computer program product as claimed in claim 20, which is executable in two parts, with a first part on the first computing means and the second part on the second computing means.
22. A computer programmed to perform the functions of the second computing means described in any one of claims 1 to 17.

lO
23. A system substantially as described herein with reference to the accompanying drawings.
24. A method substantially as described herein with reference to the accompanying drawings.