We are saddened to report that Professor Dornfeld passed away in March, 2016. If you enjoyed his blog, please consider making a contribution to either of two funds at UC-Berkeley that have been established in his memory.

David A. Dornfeld Graduate Fellowship
David A. Dornfeld Scholarship

Friday, December 3, 2010

Tools for assessing impact


Or, are we "doing the right thing?"

The last couple of postings have focussed on how to insure we can measure, and then take credit for (or get some credit for) changes made in a process that create a positive impact in terms of life-cycle impact or consumption.

This came up with reference to a discussion on net present value (or NPV) which is a way to estimate the the degree to which an improvement today leverages benefits into the future. The goal is to identify investments that can be leveraged in the future for big returns.

This brings up the question - what are some methodologies for making these assessments? In earlier postings (very early, in fact, see August, 2009 posting) we discussed rates of return for reductions in green house gas emissions, or water use, or energy use.  But, how can we identify where to apply technologies (and, more importantly, what technologies to apply) for driving these reductions?

And - apologies in advance - I've been slow to get this posting ready due to end of the year academic activities and this will be along one!

One neat technique that came to our attention for meeting this challenge is called a "pinch analysis." (see wikipedia for a good description). This came up in a recent project we were doing for a major European automotive manufacturer and the very energy intensive process they were using to clean precision components (including engine blocks and heads) after production to remove contaminants. These contaminants could lead to assembly problems and performance issues in use.

First let's look at what pinch analysis is, then the process we applied it to and then the results. (And put your thinking cap on as this will get technical fast!)

Wikipedia describes a pinch analysis as "a methodology for minimizing energy consumption" that was originally developed for the chemical industry. Wiki includes this nice summary of the technique -

"… process data is represented as a set of energy flows, or streams, as a function of heat load (kW) against temperature (deg C). These data are combined for all the streams in the plant to give composite curves, one for all hot streams (releasing heat) and one for all cold streams (requiring heat). The point of closest approach between the hot and cold composite curves is the pinch temperature (pinch point or just pinch), and is where design is most constrained. Hence, by finding this point and starting design there, the energy targets can be achieved using heat exchangers to recover heat between hot and cold streams. In practice, during the pinch analysis, cross-pinch exchanges of heat are [often] found between a stream with its temperature above the pinch and one below the pinch. Removal of those exchanges by alternative matching makes the process reach its energy target."

This example comes from the MS Thesis of Mr. Saurabh Garg, titled "Solid Particle Contaminant Cleaning in the Automotive Industry",  and done in my lab at Berkeley in Spring 2010. The motivation for this project was the large amount of energy consumed by the cleaning process that is not only a production cost constraint for the automotive industry in the wake of ever increasing energy prices, but also leads to a significant environmental footprint in terms of indirect greenhouse gas emissions. Garg noted that the severity of this impact depends on the energy mix of the geographical area and the impact created by the sources of energy production.

The objectives of the work that form the basis of applying the pinch analysis were:

- characterize various fluid flows in the process and in external circuits in terms of important parameters such as steady state flow rates, and temperature
- optimize the energy flows in the system to ensure maximum process-to-process heat recovery potential
- propose distribution of the net load on external utilities to minimize the overall heating and cooling costs, and
- analyze and compare the energy requirements of a standalone system of cleaning machines vs. that of centrally heated and cooled machines in a manufacturing assembly line.

So we are dealing with flows of fluids at different temperatures - a relatively common process characteristic in manufacturing (think painting, heat treating, washing, etc.) Not surprisingly, this will involve some simple thermodynamics.

The basic concept of a pinch analysis (as defined above) is represented by the diagram below, showing the temperature - enthalpy rate for a process stream in manufacturing. If you need some brush up on your thermodynamics, check the wikipedia discussion on enthalpy. Enthalpy is, basically, the  measure of the total energy of a thermodynamic system. Wikipedia explains that since "the total enthalpy, H, of a system cannot be measured directly … change in enthalpy, ΔH, is a more useful quantity than its absolute value. The change ΔH is positive in endothermic reactions, and negative in exothermic processes. ΔH of a system is equal to the sum of non-mechanical work done on it and the heat supplied to it."  The figure below summarizes the basis of the analysis.



The analysis starts by representing all the process streams in the domain of analysis on a temperature-enthalpy rate (T- ΔH) diagram where the vertical (y) axis represents the temperature scale while the horizontal (x) axis represents enthalpy rate. Each process stream is represented by a straight line on this diagram running from the stream inlet temperature (Tin) to the stream target temperature (Tout). For a process with a series of process streams that comprise the whole operation, you make one straight line for each stream in the series. The term ΔT stands for the difference between two temperatures.

Since any horizontal distance on the x-axis represents a difference of enthalpies in which we are interested, the absolute values on the x-axis are insignificant. It is precisely for this reason that the composite curves can be translated horizontally on a T-ΔH diagram, without affecting the process stream.  The slope of any line representing a process stream on a T-ΔH diagram is given by 1/(mass flow rate x Cp). Here Cp is the specific heat of the fluid.

For heat exchange to occur, the hot stream cooling curve (hot composite curve) must lie above the cold stream heating curve. (cold composite curve). Because of the ‘kinked’ nature of the composite curves, they approach each other most closely at one point defined as the minimum approach temperature (ΔTmin). The point of minimum temperature difference represents a bottleneck in heat recovery and is commonly referred to as “pinch” as defined earlier by the Wikipedia reference. The area of overlap between the composite curves represents the potential for process-to-process heat recovery. As stated before, horizontal translation of the curves will vary ΔTmin such that at one particular value, the overlap shows the maximum possible scope for heat recovery within the process. At this value the requirement for external hot and cold utilities, as represented by the hot and cold end overshoots of the composite curves, is minimum. However, the maximum process recovery is only a theoretical concept and practical design challenges and cost considerations limit this value as illustrated below.


As seen in the figure, external energy costs increase linearly as the ΔTmin increases. This is because at low temperature difference, the energy transfer process is more efficient and the in-process energy recovery potential is high because the hot and cold composite curves align nicely with each other. In other words, the potential for energy recovery decreases as the composite curves move apart (increasing ΔTmin).

Ok, so how was this used in the automotive cleaning example? The T-ΔH diagram depicting the hot and cold composite curves for the existing cleaning process (flows of hot and cold fluids at various temperatures) is shown in the figure below. Temperature is along the vertical axis (degrees C) and enthalpy (in kW) is along the horizontal axis. The figure was constructed following the procedure described above (and you may need to 'click' on the figure to see all the detail.)



The figure shows that there is a good potential for energy recovery through process-to-process heat exchange, as shown by the green shaded region. The pinch, in this case, is defined by an extended region and not a single position, having a minimum temperature difference of 3 degrees C.

The next step is to propose solutions to "recover" this energy and evaluate whether or not they are feasible economically and, also, what the potential environmental impact will be. A suitable heat exchanger was determined based on the area of heat exchange needed to accomplish the energy recovery. Then, using an economic analysis the potential return of the investment was determined. The figure below compares the total annual energy costs (based on heating and cooling alone) for the proposed retrofit design of the cleaning process based on an improved process-to process


heat exchange optimization vs. the current costs based on the existing design of the process. It can be seen from the figure that beyond the initial 3 years when the capital cost will be completely paid, the net difference between the operational energy costs of pinch-optimized retrofit design and the existing design is worth a savings of 84,500 Euros annually.

Further analysis resulting in considering adding a heat pump to recover some energy due to changes in fluid pressures also. That was good for another 20,000 euro savings annually after the payoff (3 years).

Finally, what about the environmental payback?

Garg includes this analysis as well. The use phase emissions for the existing cleaning process can be attributed directly to the impact created by the consumption of process electricity, and the heating and cooling energy. The total impact for each of these three forms of energy consumption can be calculated by simply multiplying the total energy requirement in each case, with a conversion factor that expresses the impact (kg CO2) per unit kWh based on the source and quality of that energy generation. For example, for a unit (kWh) electricity consumption, the corresponding GWP impact is roughly 0.649 kg CO2 equivalent based on the energy mix of Germany where this facility is located. The same is true for cooling energy, as the cooling is achieved through a refrigeration cycle that involves electricity consumption. For the heating, high temperature steam is used, whose production is linked to an equivalent impact of 0.204 kg CO2 eq./kWh.

Based on the above numbers, the use phase impact generated by the existing cleaning process is found to be 2335 MT CO2 per year. Because of the reduced energy consumption due to pinch optimization, the net impact due to the optimized process is much lower, about 1388 MT CO2 eq. per year - a "savings" of almost 1000 MT CO2 eq. per year!

However, the capital investment in the form of heat exchanger devices will also cause a one-time (fixed) impact, which can be evaluated using, for example, an Economic Input-Output Life Cycle Assessment (EIOLCA) database (e.g. from Carnegie Mellon University).  The EIO-LCA analysis for the heat exchanger was used for the given application and predicted an impact of 100 MT CO2 eq. So that is the "embedded" impact of the proposed switch and any improvement needs to be greater than that at the minimum.

Since the reduced impact, almost 1000 MT CO2 eq. per year, is substantially greater than the one time 100 MT CO2 eq. hit due to the production and installation of the heat exchanger we can safely say the GHG return on this investment is pretty good!

There is even better news. This is one cleaning station of dozens in this large automotive facility and, perhaps, hundreds throughout the company. The potential for larger impacts as more are retrofitted, with the same economic and environmental impacts, is tremendous. Talk about a great technology wedge!

And you can use this in your net present value evaluation also.

I'll let you chew on this long and detailed discussion a bit! But, the point is that there are a lot of existing tools out there that, carefully applied with solid engineering logic, can make a big impact on both bottom lines - cost and environment.

2 comments:

  1. Pinch technology has really expanded its scope over the past 15 years. What started out as energy optimisation now includes far more mass integration, especially for water and hydrogen management.

    In fact, it can be applied to any resource conservation problem where quantity and quality are important. One recent extension is to the area of time management and scheduling. More details in this article from Chemical Engineering magazine in 2010:
    http://accessintelligence.imirus.com/Mpowered/book/vche1/i5/p70

    ReplyDelete

Note: Only a member of this blog may post a comment.